GA-PSO Algorithm Research Articles

Hybrid PSO-HSA and PSO-GA algorithm for 3D path planning in autonomous UAVs

Unmanned aerial vehicles (UAVs) are a quintessential example of automation in the field of avionics. UAVs provide a platform for performing a wide variety of tasks, but in each case the concept of path planning plays an integral role. It helps to generate a pathway free of obstacles, having minimum length leading to lesser fuel consumption, lesser traversal time and helps in steering the aircraft and its corresponding antenna power signature safely around the hostile antenna to avoid detection. To optimize path planning to incorporate all the above-mentioned constraints, this paper presents two new hybrid algorithms particle swarm optimization (PSO) with harmony search algorithm and PSO with genetic algorithm. The hybrid algorithms perform both an exploratory and exploitative search, unlike the existing algorithms which are biased, towards either an exploitative search or an exploratory search. Furthermore, the hybrid algorithms are compared to the existing optimization algorithms and in all cases the hybrid algorithms give a minimum of 7% better result against PSO with up to a 40% better result against Invasive Weed optimization algorithm for a fixed computational time, suggesting better real-time applications.

Stability enhancement of a centrifugal compressor using inclined discrete cavities

Inclined discrete cavities are proposed to improve the operating stability of a centrifugal compressor while minimizing the flow loss, and optimized to simultaneously maximize the adiabatic efficiency and stall margin. Aerodynamic analysis was performed using three-dimensional Reynolds-averaged Navier–Stokes equations. The numerical results obtained for the total pressure ratio and adiabatic efficiency were validated with experimental data for the centrifugal compressor with a smooth casing. The angle of the cavity port, axial distance between cavities, and inclined angle of the cavity were selected as design variables based on sensitivity analysis. The stall margin and adiabatic efficiency at the design point were used as the objective functions for the design. Latin hypercube sampling was used to select 24 design points in the design space. Two neural network models, that is, the radial basis and deep neural networks, were tested with respect to surrogate models and their performances were compared using statistical analysis. A hybrid PSO–GA algorithm was used to identify the optimal solutions for the surrogate models. The deep neural network model has a better overall performance than the radial basis neural network model. The optimization results show that the stall margin increases, with increments of up to 6.41% and 11.32% compared with the reference design and compressor with a smooth casing, respectively.

Optimum process configuration for ETBE production based on TAC minimization

A new process configuration consisting of hybrid pervaporation (PV) (with series–parallel arrangement and recycle streams)–distillation column (DC) or reactive distillation (RD) with a hybrid feed splitting-heat integrated distillation column (HIDiC) – heat integration method was proposed to produce ETBE. To evaluate the performance of the proposed method, 92 different processes for producing ETBE by RD columns, pressure swing distillation, PV, liquid–liquid extraction column or their combinations were optimized and compared to the new configuration. The PV modules were placed in four sections: before the column, next to distillate, next to bottom product and next to side stream. All process parameters in DC or RD, PV and heat integration were considered optimization variables, and all variables were simultaneously optimized by a hybrid GA–PSO algorithm. The proposed configuration reduced TAC by at least 30% and a maximum of 62% compared with other process configurations. This can be attributed to the use of PV instead of distillation columns, series–parallel arrangement instead of series and parallel arrangements in the PV process to reduce energy consumption and membrane area and also the use of heat integration between internal DC flows and PV streams to eliminate the need for hot utility.

Flood Routing: Improving Outflow Using a New Non-linear Muskingum Model with Four Variable Parameters Coupled with PSO-GA Algorithm

Flood is one of the most destructive natural disasters that damages people’s lives dramatically. Thus, it is crucial for researchers and politicians to research flood routing. The non-linear Muskingum model has been significantly considered by engineers and researchers in flood routing. In this study, in order to increase the accuracy of outflow prediction, the new non-linear Muskingum model, with four variable parameters, is proposed for the first time. In the proposed model, the inflows are divided into three sub-regions, and each of the four hydrologic parameters has a various value in each sub-region. How to select the sub-regions, as well as the values of the hydrologic parameters, is determined by combining both the Particle Swarm Optimization and Genetic Algorithm. The proposed model is studied in four case studies. Compared to the non-linear Muskingum model with three parameters, the amount of sum squared deviation (SSQ) decreased 52 and 6.9% for the first and second case studies, respectively. Compared to the best variable parameter model, the SSQ for the third and fourth case studies reduced 76 and 62%, respectively. The results showed that the SSQ was considerably decreased significantly in all of the four case studies, and the proposed model has superiority over other non-linear Muskingum models, which have been used by other researchers so far.

Real-Time Fault-Tolerant Formation Control of Multiple WMRs Based on Hybrid GA–PSO Algorithm

A fault-tolerant formation control (FTFC) strategy is proposed against severe actuator faults, applied to a team of wheeled mobile robots (WMRs). In the beginning, a team of WMRs is operating in a prescribed formation topology. As long as the robot(s) cannot complete the required mission due to severe actuator faults, the formation is reconfigured for the healthy WMRs to eliminate the fault effects. The new reconfiguration is determined by means of an optimal assignment scheme so that each healthy robot can be assigned to a unique position. Subsequently, each robot starts planning its trajectory to reach its new position in the new formation configuration by virtue of a hybrid genetic algorithm and particle swarm optimization (GA–PSO). As metaheuristic optimization techniques, such as GA and PSO, are unable to solve the optimization problem with continuous control inputs, control parameterization and time discretization (CPTD) method is, therefore, adopted to offer an approximate piecewise linearization of the control inputs. Thus, an approach with the integration of CPTD and GA–PSO is developed. This integrated approach enables that the time of achieving the configuration is minimized, while the physical constraints of WMRs and collision avoidance are explicitly considered. Finally, real-time experiments are conducted to validate the effectiveness of the proposed algorithm compared with other optimization techniques, such as GA and PSO. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Cooperative unmanned systems have drawn significant interests in military and civilian applications. During missions’ execution, it is of great importance for cooperative unmanned systems to have fault-tolerance capabilities for achieving the desired mission when faults occur in one or more team members. A challenging problem is how to detect and isolate the fault and how to mitigate the fault effects on the whole mission. This article presents a fault-tolerant formation control strategy in the case of severe actuator fault occurrence in a team of wheeled mobile robots.

Optimization of five-parameter BRDF model based on hybrid GA-PSO algorithm

The bidirectional reflection distribution function is usually used to analyze the reflection characteristics of materials. In many cases, the BRDF models are optimized by fitting parameters. We introduced a hybrid particle swarm algorithm (GA-PSO) combining genetic algorithm and particle swarm algorithm to optimize the parameters of the five-parameter model. In order to verify the performance of the hybrid particle swarm optimization algorithm, we measured two different materials of space targets to get the experimental BRDF values. Then we simulated the parameters by using the genetic algorithm, particle swarm algorithm, and hybrid particle swarm algorithm respectively. The fitting results show that the hybrid particle swarm algorithm is better than genetic algorithm and particle swarm algorithm in accuracy and convergence speed under the same condition.

Research on Function Optimization Based on Improved Genetic Particle Swarm Optimization

Aiming at the shortcomings of single algorithm such as blind operation, directionless, long calculation time, low accuracy in Genetic Algorithm and poor diversity of population of Particle Swarm Optimization, which is prone to premature and fall into local optimum. In this study, the serial algorithm fusion idea is adopted, and the population evolved from the Genetic Algorithm is taken as the initial population to be executed by the Particle Swarm Algorithm, namely GAPSO algorithm. In view of the drawbacks of the respective algorithms in the execution of GAPSO algorithm, this paper proposes an Improved Genetic Particle Swarm Optimization algorithm, namely IGAPSO algorithm. In this algorithm, the improved GA only optimizes the initial particle once and then gives it to the improved PSO for optimization. Its optimization method and ability are greatly improved, resulting in a rapid increase in convergence speed. In addition, the improved PSO introduces adaptive inertia weight and learning factor, so that the particles can adaptively adjust global search and local search, and adaptively balance the influence of self-experience and social experience, which greatly avoids falling into local optimization problem and makes the solution more accurate. Through the verification of test functions, GAPSO has obviously improved the calculation accuracy, convergence speed and global stability compared with a single algorithm, while IGAPSO has also improved the convergence accuracy and speed compared with GAPSO, especially in multi-peak functions.

Integrated Timetable Optimization for Minimum Total Energy Consumption of an AC Railway System

The optimization of train trajectory and timetable offers two effective methods of energy-efficient operations. Most traditional research optimizes them separately and thus the global optimality cannot be achieved. Previous work has mainly concentrated on minimizing the mechanical energy consumption of trains, not the electrical energy. This paper aims to develop an integrated model to optimize the integrated timetable, which includes both the timetable and the train trajectory achieve the minimum electrical energy consumption. The AC power supply system model is first proposed to calculate the parameters of the AC power supply network. Then, a model is developed to calculate various energy consumptions using power flow calculation. Furthermore, an integrated timetable optimization model for minimum total energy consumption of AC railway system is proposed and the optimal integrated timetable is obtained by the proposed hybrid GA-PSO algorithm. Finally, 5 real-world case studies based on the Lanzhou–Xinjiang high-speed railway are presented. The results show that the proposed integrated model can achieve a reduction in total energy consumption for the entire line up to 14.3% compared with the previous optimization model. The hybrid GA-PSO algorithm achieves the best results compared with the results achieved by the GA and PSO algorithms applied along.

Optimization and heat integration of hybrid R–HIDiC–PV process with the series–parallel arrangement of PV modules and recycle streams for TAME production

Tert–amyl methyl ether (TAME) production process is consisted of a reactor, a reactive distillation (RD) column and the methanol recovery section. The bottom product of the RD column is TAME while the distillate is consisted of the azeotropic mixture of methanol and isopentane (i-pentane) that can be separated by either the pervaporation (PV) membrane or conventional pressure swing distillation (PSD) and extractive distillation (ED) methods. The PV process allows the use of both series and parallel arrangements. The series arrangement has a lower membrane area compared to the parallel arrangement, while the parallel arrangement has a lower energy cost. A series–parallel arrangement with recycle streams was used in the present study to take advantage of both the series and parallel arrangements. The HIDiC method was used for reducing energy consumption in the RD column. Moreover, heat integration between PV streams and internal flows in the RD column was used to reduce PV energy consumption. The number of modules in each row, the recycle stream, the heat exchange ratio between the rectifying and stripping sections in the RD column and the heat exchange ratio between PV streams and internal RD streams were obtained by a hybrid GA–PSO algorithm. Furthermore, all RD and PV parameters were considered optimization variables and most importantly, all process variables were optimized simultaneously. According to the results, the proposed process reduced the total annual cost (TAC) by 52% and led to a 57% energetic gain compared to the conventional hybrid RD–PSD process. The proposed process also reduced TAC by 48% and led to a 40% energetic gain in comparison with the hybrid RD–PV process. This can be related to the reduced membrane area compared to the series and parallel arrangements as well as reduced energy costs of PV heaters through heat integration between PV steams and internal flows of the column.

Two-Echelon Vehicle-Routing Problem: Optimization of Autonomous Delivery Vehicle-Assisted E-Grocery Distribution

The last mile problem of E-grocery Distribution comprises one of the most costly and highest polluting components of the supply chain in which companies deliver goods to end customers. To reduce transport cost and fuel emissions, a new element of ground-based delivery services, autonomous delivery vehicles (ADVs), is included in the E-grocery distribution system for improving delivery efficiency. Thus, the objective of this study is to optimize a two-echelon distribution network for efficient E-grocery delivery, where conventional vans serve the delivery in the first echelon and ADVs serve delivery in the second echelon. The problem is formulated as a two-echelon vehicle routing problem with mixed vehicles (2E-VRP-MV) with a nonlinear objective function, in which the total transport and emission costs are optimized. This optimization is based on the flow assignment at each echelon and to realize routing choice for both the van and ADV. A two-step clustering-based hybrid Genetic Algorithm and Particle Swarm Optimization (C-GA-PSO) algorithm is proposed to solve the problem. First, the end customers are clustered to the intermediate depots, named satellites, based on the minimized distance and maximized demand. To enhance the efficiency of resolving the 2E-VRP-MV-model, a hybrid GA-PSO algorithm is adopted to solve the vehicle routing problem. Computational results of up to 21, 32, 50, and 100 customers show the effectiveness of the methods developed here. At last, the impacts of the layout of the depot-customer and customer density on the total cost are analyzed. This study sheds light on the tactical planning of the multi-echelon sustainable E-grocery delivery network.

Application of hybrid GA-PSO based on intelligent control fuzzy system in the integrated scheduling in automated container terminal

With the development of the large ship, automated container terminals (ACTs) have serious energy consumption and carbon emission problems, reducing the loading and unloading time of ships can ease energy consumption, improve the working efficiency and service level of automated terminals. This paper studies the integrated scheduling problem of the gantry cranes (QCs), automated guided vehicles (AGVs) and automated rail-mounted gantry (ARMG) in automated terminal. According to the loading and unloading operation mode, we build the mixed integer programming model with the goal of minimizing the ship loading and unloading time, and through various algorithms of heuristic and hybrid improved to solve this problem, it proves the effectiveness of the model to obtain optimized scheduling scheme by numerical experiments, and comparing the different performance of algorithms, the results show that the hybrid GA-PSO algorithm with adaptive auto tuning is superior to other algorithms in terms of solution time and quality, which can effectively solve the problem of integrated scheduling to save the energy of automated container terminal.

Airfoil Optimization of Land-Yacht Robot Based on Hybrid PSO and GA

Airfoil optimization algorithm is studied and a hybrid PSO and GA method is proposed in this paper. After function test, it shows that algorithm is well in convergence performance, fast speed, and optimization capability. Then, the airfoil parametric expression theory is analyzed. A new airfoil is obtained after combining CFD and PSO-GA optimization. The aerodynamic of new airfoil is compared with the airfoil optimized by GA-PSO and basic airfoil NACA0018. The results indicate that new airfoil is better than the other two airfoils in lift coefficient, lift-drag ratio, and surface pressure. At last, wing-sail of new airfoil and NACA0018 wing-sail are designed and manufactured. Both of them are applied in land-yacht robot linear motion and steering motion experiment. For the linear motion, in the situation of wind speed being 15[Formula: see text]m/s and angle of attack being 5, running speed of robot with optimized new wing-sail is 1.853[Formula: see text]m/s. In steering motion, trajectory with new wing-sail is closer to the real situation and it gets more thrust. The experiments data verify that the simulation results are correct and PSO-GA algorithm is effective.

Development of an efficient hybrid GA-PSO approach applicable for well placement optimization

When it comes to the economic efficiency of oil and gas field development, finding the optimum well locations that augment an economical cost function like net present value (NPV) is of paramount importance. Well location optimization has long been a challenging problem due to the heterogeneous nature of hydrocarbon reservoirs, economic criteria, and technical uncertainties. These complexities lead to an enormous number of possible solutions that must be evaluated using an evaluation function (e.g. a simulator). This makes it necessary to develop a powerful optimization algorithm into which a fast function evaluation tool is incorporated. The present study describes the application of a combination of the genetic algorithm (GA) and the particle swarm optimization (PSO) into a hybrid GA-PSO algorithm that is implemented in a streamline simulator to determine optimal locations for production and injection wells across heterogeneous reservoir models. Performance of the hybrid GA-PSO algorithm is then compared to that of the PSO and the GA separately. The results confirm that compared to conventional methods, the recommended method provides a fast and well-defined approach for production optimization complications. Cited as : Yazdanpanah, A., Rezaei, A., Mahdiyar, H., Kalantariasl, A. Development of an efficient hybrid GA-PSO approach applicable for well. Advances in Geo-Energy Research, 2019, 3(4): 365-374, doi: 10.26804/ager.2019.04.03

A Hybrid PSO-GA Algorithm for Traveling Salesman Problems in Different Environments

In this study particle swarm optimization (PSO) is modified and hybridised with genetic algorithm (GA) using one’s output as the other's input to solve Traveling Salesman Problem(TSP). Here multiple velocity update rules are introduced to modify the PSO and at the time of the movement of a solution, one rule is selected depending on its performances using roulette wheel selection process. Each velocity update rule and the corresponding solution update rule are defined using swap sequence (SS) and swap operation (SO). K-Opt operation is applied in a regular interval of iterations for the movement of any stagnant solution. GA is applied on the final output swarm of the PSO to search the optimal path of the large size TSPs. Roulette wheel selection process, multi-point cyclic crossover and the K-opt operation for the mutation are used in the GA phase. The algorithm is tested in crisp environment using different size benchmark test problems available in the TSPLIB. In the crisp environment the algorithm gives approximately 100% success rate for the test problems up to considerably large sizes. Efficiency of the algorithm is tested with some other existing algorithms in the literature using Friedman test. Some approaches are incorporated with this algorithm for finding solutions of the TSPs in imprecise (fuzzy/rough) environment. Imprecise problems are generated from the crisp problems randomly, solved and obtained results are discussed. It is observed that the performance of the proposed algorithm is better compared to the some other algorithms in the existing literature with respect to the accuracy and the consistency for the symmetric TSPs as well as the Asymmetric TSPs.

Economic Load Dispatch in Power System by Hybrid Swarm Intelligence

Conserving energy and efficient power systems are very important for us to reduce pollution levels and also reduces wasted fuel resources which are already depleting on the planet. In power operation system, the potential of energy conservation and also much less emission of greenhouse gas because of the wise usage of cleaner non-renewable fuels burned in combined heat and power (CHP) models like the natural gas which provide them benefit from the usual electric power systems. Mixed generator systems have been widely employed by the industry. The industry which requires both power and heat can supply the demands with cogeneration of heat-power systems. Co-generation (CHP) systems could be constructed in cities and used in the form of distributed electricity sources. To get the optimal usage of CHP devices, economic load dispatch (ED) should be requested more for the process of energy conservation. Economic load dispatch plays a vital role and a large number of different approaches and methods have been used in solving such kind of problems. The methods like lambda-iteration and Gradient are used for finding out the optimized solution of non-linear problem. The purpose of this thesis is to utilize the algorithmic optimization approaches like particle swarm optimization (PSO), genetic algorithm (GA) and PSO-GA. In this work, the method of PSO-GA Optimization is used to find out the minimized cost at four of the generating units of heat and power. The base of the work is already published where in the loss coefficients are also presented with max-min cost function and power limit. This work is implemented in the MATLAB simulation environment. The work starts by initializing the load/power and then generators load power flow. The generators are allocated to initialize the cost and this cost is optimized by PSO with GA. At the end the experimental results of GA ,PSO and PSO-GA algorithm is equated with each other and it seems better convergence is achieved by PSO-GA Algorithm.

An Intelligent Optimization System for PIM Process

This study proposes an intelligent optimization system based on the Taguchi method, back-propagation neural network (BPNN), multilayer perceptron (MLP) and modified PSO-GA to find optimal process parameters in plastic injection molding (PIM). Firstly, the Taguchi method is used to determine the initial combination of parameter settings by calculating the signal-to-noise (S/N) ratios from the experimental data. Significant factors are determined using analysis of variance (ANOVA). The S/N ratio predictors (BPNNS/N) and quality predictors (BPNNQ) are constructed using BPNN with the experimental data. In addition, a modified PSO-GA algorithm in conjunction with MLP is used to find initial weights of BPNN and to reduce the training time of BPNN. In the first stage optimization, the S/N ratio predictors are coupled with GA to reduce the variations of the manufacturing process. In the second stage optimization, The combination of S/N ratio predictors and quality predictors with modified PSO-GA is empoyed to search for the optimal parameters. Finally, three confirmation experiments are performed to assess the effectiveness of these approaches. The experimental results show that the proposed system can create the best performance, and optimal process parameter settings which not only enhance the stability in the whole injection molding process but also effectively improve the PIM product quality. Furthermore, experiences of the novel hybrid optimization system can be transferred into the intelligent PIM machines for the coming up internet of things (IoT) and big data environment.

Applying multiple decomposition methods and optimization techniques for achieving optimal cost in mixed materials heat exchanger networks

The optimization of the total annual cost in heat exchanger networks has been one of the overarching goals when synthesizing these networks. Several methodologies and techniques have been developed to achieve optimal costs in mixed material heat exchanger networks. This paper demonstrates the application of two decomposition methodologies (total decomposition and partial decomposition) for typical cost rules. The objective function was defined as the optimization and minimization of the total annual cost in mixed materials heat exchanger network. Three optimization algorithms, hybrid genetic-particle swarm optimization (GA-PSO), shuffled frog leaping algorithm (SFLA) techniques, and ant colony optimization (ACO), were used to further optimize the total cost in mixed materials heat exchanger network. The results indicate that the total annual cost in partial decomposition method was smaller than that in full integration method and total decomposition method. The reduction of the total annual cost was about 27% for GA-PSO algorithm, 24% for SFLA and 10% for ACO relative to the results reported in this work. In partial decomposition method, at least one mixed material of heat exchanger was used to reduce the hot and cold utility for decreasing the total annual cost. Partial decomposition method resulted in the highest reduction of the total annual cost compared with other methods. Percentage of difference of the total annual cost were 0.36%, 1.92%, and 5.05% for full integration, total decomposition, and partial decomposition methods, respectively, in comparison with the previous studies. Results have been compared with the results of other studies to demonstrate the accuracy of the applied algorithms.

Optimization study of thermal-storage PV-CSP integrated system based on GA-PSO algorithm

In thermal-storage photovoltaic-concentrated solar power (PV-CSP) systems, the fluctuant part electricity is stored in thermal energy storage (TES) system instead of high-cost batteries. In this research, PV nominal power, system power output point and TES capacity in different dispatch strategies are optimized by GA-PSO to achieve a minimized levelized cost of electricity. Then, more power output points are set in each strategy, operating performance and optimal configuration in these complex strategies are studied. After that, optimal configurations and operating characteristics in different heliostats and PV panels cost are explored, and the battery bank is considered. Results show that for the selected CSP subsystem in constant-output strategy, in which CSP power responds to PV power output to provide a constant system power curve, the optimal PV nominal power is 35.23 MW, system power output point is at the CSP-rated power and LCOE is 16.93 ¢/kWh. For independent-output strategy, in which CSP and PV provide power independently, optimal PV nominal power is 213.91 MW, PV power output point is 76.10 MW and LCOE is 16.33 ¢/kWh. For constant-output strategy with one (50% load) or three (75%, 50%, 25% load) middle power output points settled, the optimal systems based on these complex strategies can reach a lower LCOE of 15.32 ¢/kWh and 15.15 ¢/kWh, respectively. In independent-output strategy, LCOE of optimal systems based on complex strategies is 14.33 ¢/kWh and 12.80 ¢/kWh, respectively. In general, the operating performance are more superior and steadier for CON strategy and a large-scale PV subsystem is needed for IND strategy in most conditions.

Comparison of Optimal Hedging Policies for Hydropower Reservoir System Operation

Reservoir operation rules play an important role in regions economic development. Meanwhile, hedging policies are mostly applied for municipal, industrial, and irrigation water supplies from reservoirs and it is less used for reservoir operation for hydropower generation. The concept of hedging and rationing factors can be used to maintain the water in a reservoir for the sake of increasing water storage and water head for future use. However, water storage and head are the key factors in operation of reservoir systems for hydropower generation. This study investigates the applicability of seven competing hedging policies including four customary forms of hedging (1PHP, 2PHP, 3PHP, DHP) and three new forms of hedging rules (SOPHP, BSOPHP, SHPHP) for reservoir operation for hydropower generation. The models were constructed in MATLAB R2011b based on the characteristics of the Batang Padang hydropower reservoir system, Malaysia. In order to maximize the output of power generation in operational periods (2003–2009), three optimization algorithms namely particle swarm optimization (PSO), genetic algorithm (GA), and hybrid PSO-GA were linked to one of the constructed model (1PHP as a test) to find the most effective algorithm. Since the results demonstrated the superiority of the hybrid PSO-GA algorithm compared to either PSO or GA, the hybrid PSO-GA were linked to each constructed model in order to find the optimal decision variables of each model. The proposed methodology was validated using monthly data from 2010–2012. The results showed that there are no significant difference between the output of monthly mean power generation during 2003–2009 and 2010–2012.The results declared that by applying the proposed policies, the output of power generation could increase by 13% with respect to the historical management. Moreover, the discrepancies between mean power generations from highest to lowest months were reduced from 49 MW to 26 MW, which is almost half. This means that hedging policies could efficiently distribute the water-supply and power-supply in the operational period and increase the stability of the system. Among the studied hedging policies, SHPHP is the most convenient policy for hydropower reservoir operation and gave the best result.

Short term electric load forecasting model and its verification for process industrial enterprises based on hybrid GA-PSO-BPNN algorithm—A case study of papermaking process

Process industry consumes tremendous amounts of electricity for production. Electric load forecasting could be conducive to managing the electricity consumption, determining the optimal production scheduling, and planning the maintenance schedule, which could improve the energy efficiency and reduce the production cost. This paper proposed a short term electric load forecasting model based on the hybrid GA-PSO-BPNN algorithm. The GA-PSO algorithm is used in a short-term electric load forecasting model to optimize the parameters of BPNN. The forecasting model avoids the shortcoming that the prediction result is easy to fall into local optimum. The papermaking process, as one of the most representative process industries, is selected as the study case. The real-time production data from two different papermaking enterprises is collected to verify the proposed model. Besides the proposed GA-PSO-BPNN model, the GA-BPNN and PSO-BPNN based electric load forecasting models are also studied as the contrasting cases. The verification results reveal that the GA-PSO-BPNN model is superior to the other two hybrid forecasting models for future application in the papermaking process since its MAPE is only 0.77%.