Energy-saving Performance Research Articles

Health-conscious predictive energy management strategy with hybrid speed predictor for plug-in hybrid electric vehicles: Investigating the impact of battery electro-thermal-aging models

Improving plug-in hybrid electric vehicles (PHEVs) fuel economy requires a proper energy management strategy (EMS). Efforts to enhance the energy-saving performance of predictive EMSs have concentrated on advanced speed prediction methods. However, the impact of battery models on the predictive EMS hasn't been investigated. This paper aims to fill the research gap by suggesting a model predictive control-based (MPC) EMS framework that uses a hybrid speed predictor. Firstly, six control-oriented battery electro-thermal-aging models with various terminal voltage and temperature simulation accuracies have been developed and verified. Secondly, it also examines the effects of different battery models on MPC-based EMS through quantitative analysis, including battery dynamics, calculation time, and resultant operating costs, which leads to the suggestions of model selection for the design of the EMS under low- and room-temperature driving scenarios. Finally, a novel hybrid speed prediction model is proposed, where the historical speed is decomposed into strongly periodic intrinsic mode functions (IMFs) by variational mode decomposition (VMD), and backpropagation (BP) neural network is utilized to learn the feature parameter and mapping relationship of each IMF. In addition, a case study is conducted by applying the proposed speed prediction model in an MPC-based EMS method. The simulation results highlight that the proposed hybrid speed prediction model can achieve preferable speed prediction. The operating cost errors (compared with the MPC-based EMS with 100% speed prediction) are reduced to 0.4% and 0.98% at −20 °C and 25 °C driving scenarios, respectively.

Joint optimization of energy conservation and transfer passenger service quality in rail transit system

ABSTRACT The complex electricity transmission process and transfer passenger demand in the rail transit system make it challenging to formulate optimal timetables and train speed profiles to meet the needs of the system. This study optimizes train timetable to meet energy conservation and passenger service quality of cross-mode rail transit system during different operation periods. Firstly, based on the spatiotemporal travel data of transfer passengers, the train-time matrix is proposed to construct a bi-objective optimization model (BOOM) that incorporates the total energy consumption optimization model (TECOM) and total waiting time optimization model of transfer passenger (TWTOM). It can better combine the passenger transport scheme and the train energy-saving operation strategy, thus improving energy-saving performance and passenger transport efficiency and simplifying the difficulty of mathematical modelling of such BOOM. Then, a method for determining the appropriate weight coefficient ratio and the improved non-dominated sorting genetic algorithm II (NSGA-II) are applied to obtain the efficient Pareto front solutions. Taking the numerical experiments as test cases, the improved NSGA-II can get a efficient solution faster than the traditional NSGA-II. Finally, two numerical experiments using real-world data are conducted to verify the practicability of the optimization model and the effectiveness of improved NSGA-II. The optimization results can reduce energy consumption by up to 11.02% and passenger waiting time by 137,320 s, respectively. This study helps to improve the energy efficiency and passenger service quality of cross-mode rail transit systems, and also provides a reference for dispatchers to optimize train timetable.

Performance characteristics of a compact self-circulating liquid desiccant air dehumidification system coupled with a heat pump: A comprehensive parametric study

Integration of a heat pump with a liquid desiccant system for air dehumidification has received growing attention in recent years in view of its more compact size and lower electricity consumption. This paper presents a compact combined heat pump and self-circulating liquid desiccant air dehumidification system targeted at small cooling load applications. The compactness lies in the integration of the dehumidifier and evaporator as well as the integration of the regenerator and condenser. Besides, the design of solution self-circulation (SSC) is intended for energy savings and the auxiliaries are employed to help achieve energy matching between the heat pump and the air dehumidification system. First, a comparative study is performed to identify the more energy efficient one of the two strategies for energy matching. Based on this, a thermodynamic model-based parametric study is conducted to investigate the effects of various influencing factors on the energy matching relationship and performance of the proposed system, as well as the energy-saving potential relative to the basic system without SSC. The results show that there exists an optimum solution self-circulation ratio (SCR) to achieve the maximum COP under a given condition. Basically, the proposed system with a proper SCR could yield a significant energy-saving effect regardless of the performance variations under studied operation conditions. It is found that a reduction of 29.68 % in energy consumption and an improvement of 42.20 % in COP are achieved under the ambient condition of 30 ℃ temperature and 20 g/kg humidity ratio. Furthermore, a correlation analysis is used to reveal the association strength and direction between any two main operation and performance parameters. Finally, the heat recovery is proved to effectively reduce the cooling coils load. This paper provides an in-depth understanding of the energy matching relationship and demonstrates the energy-saving performance benefiting from SSC for future practical applications of the proposed system.

Design and performance investigation of a novel self-adaptive radiative cooling module for thermal regulation in buildings

Daytime radiative cooling utilizes the process of reflecting sunlight and radiating heat through the atmospheric transparent window (8–13 μm) to achieve spontaneous cooling of an object. However, the existing challenge lies in the disparity between the cooling supply and demand, necessitating the development of self-regulating radiative cooling. This study introduces a novel design that combines paraffin wax with a radiative cooler, leading to the development of a self-adaptive radiative cooler (SRC) with two distinct modes based on dynamic optical properties. The spectral properties of the four SRCs were calculated by Fresnel equation, and it was found that the SRC possessed the ability to optimally absorb solar energy (Δα = 0.322) and automatically adjusted their thermal emittance (Δε = 0.552) in response to ambient temperature changes, facilitated by the liquid-solid transition, especially for Case 1. Therefore, the novel SRCs demonstrate a unique behaviour: they are warmer than static radiative coolers (ΔT = 3 K) in cold ambient conditions, while maintaining high cooling power in hot environments. Through extensive simulations for various cities and climates, this study demonstrates the superior energy-saving performance of the SRCs in building thermal regulation compared to that of static radiative coolers (Δδ1 = 9.6%).

Comparative study of waste heat recovery systems based on thermally driven refrigeration cycles in cryogenic air separation units

Recovering compression heat generated in the air compression process is expected for energy saving of cryogenic air separation units. However, the effective utilization of compression heat remains a challenge due to the low-grade feature of the available heat source. Therefore, the selection of the waste heat recovery system configuration is critical to the optimum utilization of the compression heat. In this study, two waste heat recovery systems based on different thermally driven refrigeration cycles for compression heat recovery are proposed, including an organic Rankine vapor compression-based air compression system (ORVC-ACS) and an absorption refrigeration-based air compression system (ARS-ACS). A 60,000-Nm3/h scale cryogenic air separation unit is considered as a case study for assessing and comparing the energetic, exergy and economic potential of the two proposed systems. The results reveal that the ORVC-ACS is preferable to the ARS-ACS in terms of energy-saving and economic performance. Operation optimization is conducted to identify the maximum energy saving of the preferred system. The optimization results show that the ORVC-ACS is capable of saving a total electricity of 8.52 GWh/year, corresponding to 4.53% of the annual electricity consumption for air compressors, while the values are 6.43 GWh/year and 3.37% for the ARS-ACS. This work may provide guidelines for configuration selection and optimum design of waste heat recovery systems in cryogenic air separation units.

A health-aware energy management strategy for fuel cell hybrid electric UAVs based on safe reinforcement learning

Energy management strategies (EMSs) are crucial for the hydrogen economy and energy component lifetimes of fuel cell hybrid electric unmanned aerial vehicles (UAVs). Reinforcement learning (RL)-based schemes have been a hotspot for EMSs, but most of RL-based EMSs focus on the energy-saving performance and rarely consider energy component durability and safe exploration. This paper proposes a health-aware energy management strategy based on a safe RL framework to minimize the overall flight cost and achieve safe operation of UAVs. In this framework, a universal three-dimensional environment that integrates the UAV kinematics and dynamics model is developed. In addition, wind disturbances and random loading of the mission payload during flight are considered for robust training. The energy management problem is formulated as a constrained Markov decision process, where both hydrogen consumption and energy component degradation are incorporated in the multi-objective reward function. A safety optimizer is then designed to satisfy operation constraints by correcting the action through analytical optimization. The results indicate that the safety of the explored action is guaranteed, maintaining zero constraint violations in both training and real-time control scenarios. Compared with other RL-based methods, the proposed method had better convergence capability and reduced the training time. Furthermore, the simulation showed that the proposed method can reduce the total flight cost and fuel cell degradation by 14.6% and 15.3%, respectively, compared with the online benchmark method.

Research on the cooperative train control method in the metro system for energy saving

PurposeTo address the problem that the current train operation mode that train selects one of several offline pre-generated control schemes before the departure and operates following the scheme after the departure, energy-saving performance of the whole metro system cannot be guaranteed.Design/methodology/approachA cooperative train control framework is formulated to regulate a novel train operation mode. The classic train four-phase control strategy is improved for generating specific energy-efficient control schemes for each train. An improved brute force (BF) algorithm with a two-layer searching idea is designed to solve the optimisation model of energy-efficient train control schemes.FindingsCase studies on the actual metro line in Guangzhou, China verify the effectiveness of the proposed train control methods compared with four-phase control strategy under different kinds of train operation scenarios and calculation parameters. The verification on the computation efficiency as well as accuracy of the proposed algorithm indicates that it meets the requirement of online optimisation.Originality/valueMost existing studies optimised energy-efficient train timetable or train control strategies through an offline process, which has a defect in coping with the disturbance or delays effectively and promptly during real-time train operation. This paper studies an online optimisation of cooperative train control based on the rolling optimisation idea, where energy-efficient train operation can be realised once train running time is determined, thus mitigating the impact of unpredictable operation situations on the energy-saving performance of trains.

Evaluation of fuel consumption and emissions benefits of connected and automated vehicles in mixed traffic flow

Introduction: The presence of connected and automated vehicles (CAV) in mixed traffic flows with different market penetration rates (MPRs) in urban road scenarios has a significant effect on fuel consumption and exhaust emissions.Methods: Therefore, in this study, real-world road networks and traffic data are simulated using SUMO based on actual data from a survey. The fuel consumption and emission benefits of CAVs in mixed traffic flows are well-evaluated, and the energy-saving performance of CAVs under low-speed vehicle interference is tested. In addition, this study explores both the energy consumption and emissions of purely electric vehicles.Results: The results show that with 100% CAV penetration, fuel vehicles have a maximum reduction in fuel consumption of 18% and a maximum increase in average speed of 31.6%, while the energy consumption of electric vehicles increases due to communication, detection, and collaboration between CAVs.Discussion: However, the results clearly demonstrate that the carbon emissions of electric vehicles are significantly lower than fuel vehicles. In addition, the increase in low-speed vehicles will result in an increase in energy consumption and emissions. Therefore, increasing the percentage of electric vehicles on the roads and transitioning from manual to autonomous driving systems is crucial to curbing carbon emissions.

Integrated optimal control of distributed in-wheel motor drive electric vehicle in consideration of the stability and economy

From the perspective of handling stability and economy, this paper brings forward an integrated optimal control scheme of distributed in-wheel motor drive electric vehicles (IWMDEV). A linear quadratic regulator (LQR) controller is presented according to nominal model and ideal motions firstly. To tackle the uncertain parameters and random disturbance, a novel robust optimal (RO) control strategy is further proposed to design the top level controller by employing the combination of optimal control and high order sliding mode control. By virtue of the integral and terminal sliding mode surface, the resulting robust optimal control method can not only achieve better performance, but also converge in finite time without chattering phenomenon. As for the bottom level controller, an optimal algorithm is developed to determine the weights for each term in objective function of torque allocation dynamically and coordinate the relationship between driving safety and energy-saving performance timely. Finally, joint simulations are carried out. In comparison with LQR control strategy, the results demonstrate that the proposed control scheme could ensure the robustness and effectiveness in the presence of parametric perturbation and external disturbance. Moreover, the torque allocation algorithm, which considers both handling stability and energy consumption, proves to be an effectual approach.

Design and Optimization of Flywheel Energy Storage System for Rail Transit

At present, the urban rail transit system has problems such as energy waste in the braking process and unstable grid voltage in the start-stop state. Aiming at the problems caused by the start-stop state of rail transit, considering the energy saving and voltage stability requirements of system energy management, a flywheel energy storage system (FESS) specially used for rail transit is designed. The energy system (FESS) can feed back the braking energy stored by the flywheel to the urban rail train power system when the rail train starts to cause the voltage and frequency of the traction microgrid to change. This paper proposes a flywheel energy management system based on a permanent magnet synchronous motor (PMSM), which can realize efficient energy management through efficient control of the flywheel side motor. The flywheel side permanent magnet synchronous motor adopts an improved flywheel speed expansion energy storage control strategy based on current feedforward control to improve the fast response capability of the flywheel, ensuring the output power of the flywheel at high speed. In addition, a specific multi-threshold voltage single FESS control strategy is suggested, which improves the energy-saving and voltage-stabilizing performance of the system.

Toward a Finite-Time Energy-Saving Robust Control Method for Active Suspension Systems: Exploiting Beneficial State-Coupling, Disturbance, and Nonlinearities

A novel control method of addressing coupling and disturbance influences for finite-time energy-saving robust control of active suspension systems (ASSs) is investigated. By elaborately constructing coupling and disturbance effect indicators, the pros and cons of coupling and disturbance influences on ASSs are discussed, and then a finite-time coupling and disturbance effects-triggered control method is designed via a second-order sliding mode control technique. Importantly, the good/bad coupling effects are assessed through a well-designed nonlinear function. By means of determining if the sign of disturbances conforms to the expected motion or not, the addition of beneficial disturbance effects or removal of detrimental disturbance effects is implemented. Noticeably, by employing a bioinspired nonlinear reference model, beneficial nonlinear stiffness and damping effects are thus utilized, leading to the possibility of energy-saving performance. As a result, the proposed control method exhibits a unique feature, i.e., fully employing potential contribution from the coupling and disturbance effects, and presents a totally new coupling and disturbance effects-triggered control framework, leading to obvious performance improvement. Benchmark experimental conclusions are devoted to distinguishing the advantages and effectiveness of the designed tracking method.

Distributed multi-heat-source hybrid heating system based on waste heat recovery for electric vehicles

Air-conditioners usually consume the most electricity among all of the auxiliary components in an electric bus, over 30% of the battery power at maximum. A distributed multi-heat-source phased control method was developed to reduce the energy consumption of heating in electric vehicles (EVs). This method combines the application of multiple heat sources, including waste heat from battery cooling, motor and controller cooling, and positive temperature coefficient heating. Analyzed the internal thermal process and heat generation law of the power battery, and explored the thermal process of the driving motor and its controller. Apply the cooling waste heat from the power battery and the driving motor to the heating system of pure electric vehicles. The heat sources are distributed at different locations on the vehicle and provide heating in stages at various temperatures. Numerical analysis was conducted to calculate the heat output of each heat source, and a phased control strategy for the air conditioning system was designed according to the heat release characteristics and magnitudes of each component. Heating control functions for different stages were established at varying temperatures. The temperature rise patterns at different points on the vehicle were examined, and the positions of the heat sources were redesigned accordingly. The proposed method selects the appropriate heating mode according to factors such as the external temperature, the heat release sequence of the heating components, and passenger comfort requirements. Low-temperature experiments were performed at different temperature stages to compare the heating performance of this method with those of traditional heating methods and heat pump air conditioning for new energy vehicles. The experimental results show that the heating system has good heating effect. Compared with heat pump air conditioning and PTC heating, the proposed collaborative system saved 31.1% and 63.6% of energy, respectively, after operating at − 22℃ for 2 h. It demonstrates good energy-saving and heating performances at different temperature stages. The heating method is currently one of the best energy-saving solutions for EVs.

Thermal performance of vertical greenery systems (VGS) in a street canyon: A real-scale long-term experiment

Outdoor vertical greenery systems’ (VGS) thermal aspects have been intensively studied in recent years; nevertheless, real-scale long-term experiments are scarce in the literature, and little is known about the thermal effect of VGS in street canyons. An experiment on east–west-oriented real buildings in street canyons was conducted for 31 months, analyzing the thermal performance of two VGS technologies using sensors and a thermal camera. Energy saving was assessed and an allometric model was used for carbon sequestration estimation. We present time series analyses of the diurnal and seasonal cycles of the outdoor and indoor environments with respect to bare walls street canyon. Daytime cooling in the middle of the canyon 10 m from the VGS persisted for 59.4% of the time in summer and 79.4% in a heatwave, while heating persisted for 73.9% of the day in the winter. The cooling effect was more efficient in higher temperatures and was stronger 1 m near the VGS. The surface temperature of the sunlit VGS was cooler by an average of −3.8 °C in the summer and −6.8 °C in a heatwave and was slightly warmer during the winter. The average indoor summer cooling effects were greater on the south-facing orientation (air −1.57 °C; wall −1.31 °C), while the winter warming effect was greatest on the north-facing orientation (air 2.67 °C; wall 1.6 °C). The yearly energy-saving performance due to air conditioning was 8.9%. The CO2 sequestration potential (140 kg/year) was negligible compared to the CO2 emissions that were saved from air conditioning (∼9 tons/year).

Hydrothermal and energy-saving performances in a mini-channel heat sink under different wire coil inserts induced swirling flow

Energy conservation and emission reduction are crucial international policies. For micro/mini-channel heat exchangers, improving their thermal performances while consuming as little energy as possible is essential for implementing these policies. Swirling flow stands out as one of the most effective methods for heat transfer enhancement in channel. The wire coil insert, known as an effective swirling flow device, has been widely applied in traditional heat exchangers. However, the effects of different wire coil inserts on hydrothermal and energy-saving performances in the square micro/mini-channel remain unclear. Actually, some wonderful swirling flow generators achieve excellent thermal performance at the cost of sacrificing flow performance and energy conservation. This violates the primary intention of energy conservation and emission reduction. Therefore, this study designs several wire coil inserts with various shapes (square, circle, obverse or reverse trapezoidal, observe or reserve triangular, and hexagonal cross-section shapes) and different pitches (p=3–10.5 mm). Also, a numerical investigation on effects of these wire coil inserts on performances in mini-channels (a single mini-channel with 2×2 mm2 cross-section) at Reynolds number (Re) of 523–1120 is conducted. The results show that the square wire coil insert yields the best heat transfer and comprehensive thermal performance. This is attributed to the fact that wire coil inserts induce swirling flow, which promotes the separation and redevelopment of the flow and thermal boundary layers. Based on this, the effect of wire coil pitch is analyzed thoroughly. Increasing the pitch leads to an increasing first and then decreasing tendency in the comprehensive thermal performance. Overall, the mini-channel with a square wire coil insert at p=6 mm obtains the optimal comprehensive thermal performance factor (∼2.0) for Re = 1120. Moreover, it also presents an acceptable energy-saving performance. In addition, the comparison of this study and several other studies suggests that, square coil insert has great potential to improve comprehensive performance of mini-channel heat exchanger at higher Reynolds number.

A multi-strategy improved sparrow search algorithm of large-scale refrigeration system: Optimal loading distribution of chillers

At present, the energy consumption of the parallel chillers system accounts for about 60% of the whole energy consumption of central air conditioning system and 25– 40% of the total energy consumption. Unreasonable load distribution brings huge energy consumption and carbon emissions during building operation. In order to reduce energy consumption and carbon emissions of the parallel chillers system in large-scale refrigeration system, we proposed an improved sparrow search algorithm (TMSSA). This algorithm initializes the population using Tent chaotic map to increase population diversity, which can enhance the possibility of TMSSA to obtain the optimal solution before the optimization iteration, and speed up the convergence process. Besides, a Levy flight mechanism is applied to improve the position update of the producer, enhancing the randomness and local search ability of this algorithm. Moreover, Gaussian mutation method is utilized to perturb the position of scroungers, strengthening the ability of the algorithm to escape local optima to improve robustness of TMSSA. To evaluate the proposed algorithm, 12 benchmark functions were chosen, and the results showed that it overcomes the limitations of traditional SSA regarding local optima trapping. The algorithm is also used to solve the optimal chiller loading (OCL) problem, which has 3 typical cases, and comparisons were made with other algorithms. The results further demonstrate that TMSSA is highly accurate, fast-converging, and robust with excellent energy-saving performance.

Enhancing energy efficiency of chemical absorption-based CO2 capture process with advanced waste-heat recovery modules at a high capture rate

A commercial-ready amine-based CO2 capture process (CCP) restricts its potential for industrial deployment at a high capture rate to achieve a net-zero carbon target due to its energy-intensive nature. This study aims to enhance the energy efficiency of a monoethanolamine (MEA)-based CCP, operating at a 95% capture rate, by developing advanced process modules that utilize low-grade waste heat. To establish a foundation for designing cascade modules that can recover energy from all available waste heat, the waste heat characteristics were analyzed, and the feasibility of four single modules was explored to recover the latent heat in a stripped gas. Subsequently, this study comparatively investigated the equivalent electrical works of the two designed cascade modules combined with lean vapor compression (LVC). A techno-analysis indicated that cascade modules provided outstanding energy-saving performance ranging from 12.27% to 24.55%, with an average improvement of 50% over single modules. Although the cascade modules combined with LVC showed similar energy-saving performances, higher robustness for equivalence factor (EF) and more energy-savings were discovered in the electricity generation and direct heat supply strategies, respectively. Economic analysis indicated that advanced process modules were economically feasible, with a savings-to-investment ratio (SIR) exceeding 1. The SIR and payback period (PBP) depended on EF and electricity prices. Significantly, the cascade module for direct heat supply emerged as the most appropriate in terms of both economic viability and energy-saving efficiency in amine-based CCPs, particularly at a higher EF.

The performance investigation and optimization of reciprocating flow applied for liquid-cooling-based battery thermal management system

Liquid cooling is the most popular battery thermal management system (BTMS) at present, while suffers from high energy consumption and high temperature difference between upstream and downstream. Herein, we first use a reciprocating liquid flow-based BTMS (RLF-BTMS) for cylindrical batteries to release those issues. The comparison among unidirectional flow, cross-direction flow, and reciprocating flow was carried out to investigate the effectiveness of reciprocating flow. After that, the effects of velocity and reciprocating period, two key factors of reciprocating flow, were systematically explored. Finally, a comprehensive criterion of the performance of BTMS considering maximum temperature, temperature difference, temperature distribution coefficient, and energy consumption was proposed to optimize the reciprocating flow. The temperature difference and energy consumption are reduced by 55.3 % and 15.6 % through reciprocating flow. To get energy-saving performance, the reciprocating flow velocity ought to be lower than 0.085 m/s. Since has been greatly reduced, temperature difference changes insensitively with velocity. As decreased from unidirectional flow to 0, the reciprocating periods are divided into three stages: disorder stage, positive stage, and worse stage, according to the thermal characters. Based on the novel BTMS evaluation index, 0.05 m/s and 200 s are selected as the optimal velocity-period combination for the RLF-BTMS. This study recommends applying reciprocating flow to replace the traditional unidirectional flow to get better cooling performance.

A Data-Driven Iterative Learning Approach for Optimizing the Train Control Strategy

The energy-efficient train control (EETC) problem is investigated in this paper. And a soft actor-critic-based method is proposed to optimize the train driving strategy. Firstly, EETC problem is converted to the inverse problem, i.e., minimizing the trip time of the journey with constant energy consumption. Based on the conversion, the EETC problem is reformulated as a finite Markov decision process which can be solved by deep reinforcement learning algorithms. Secondly, an optimization method based on the soft actor-critic (SAC) method is designed to calculate the optimal driving strategy of the train with introducing the reservoir sampling method. Finally, some case studies are conducted to verify the effectiveness and performance of the proposed method. Simulation results demonstrate that a good energy-saving performance can be achieved. In single interval, the SAC-based method can reduce about 1.65% of the energy consumption compared with numerical method. And the energy consumption reduction can be extended to be 6.49% when the proposed approach is applied in multiple intervals.

Research on Energy Savings of an Air-Source Heat Pump Hot Water System in a College Student’s Dormitory Building

Centralized hot water systems are commonly installed in college student dormitories, representing a typical application for such systems. To achieve sustainable and environmentally friendly heating solutions, air-source heat pump hot water systems have gained attention for their high efficiency and energy-saving characteristics. By implementing heat pump technology, China could make significant progress towards achieving its carbon neutrality goals by reducing energy consumption and associated emissions. In this study, the heating performance of an air-source heat pump hot water system was tested in the field over the course of a year at a university in Chongqing, China. A simulation model was constructed using TRNSYS software, and a time-sharing control strategy was proposed to analyze the system’s operating characteristics and energy-saving performance. Results showed a 6% increase in unit annual average Coefficient of Performance (COP) and annual electricity savings of 10,027 kW·h, with an energy-saving rate of 8.77% after time-sharing control. The study highlights the significant economic and environmental benefits of adopting sustainable energy solutions, particularly in the context of increasing global greenhouse gas emissions.

Predictive control optimization of chiller plants based on deep reinforcement learning

The energy consumption of HVAC systems is enormous, with chiller plants accounting for more than 50% of it. To improve energy efficiency, chiller systems are typically optimized at fixed intervals based on real-time building cooling loads, which usually assumes that the cooling load remains constant within the control interval. However, in many real applications, the current optimal control may be changed when the cooling load suddenly fluctuates. Although this issue can be addressed by shortening the control intervals, more frequent control can damage the equipment or increase energy consumption. To tackle this problem, this paper proposes a model-free predictive control method based on Reinforcement Learning (RL) control and Long Short-Term Memory (LSTM) prediction networks. The LSTM network aims to predict the future cooling load based on historical cooling load data, while the RL is used to make the best control for all chiller plants. By this way, the chilled water supply temperature setpoints can be optimized with the consideration of both instantaneous and predicted cooling loads to minimize energy consumption. In order to validate the effectiveness of the proposed method, an experimental simulation model was constructed based on actual chiller system parameters. Experimental results show that the energy-saving performance of the proposed method is superior to rule-based control (9% improvement), closely comparable to the model predictive control (0.73% difference), and further enhances energy-saving effects compared to non-predictive RL control without shortening the control interval. Additionally, the proposed method just learns by continuously interacting with the environment and does not require any accurate equipment models, making it a viable alternative for buildings when lacking extensive sensors and device information.