Variable Working Conditions Research Articles

Study on circulating cooling water pre-mixing for thermal performance improvement of twin thermal power units sharing one dry cooling tower

For an advanced power plant of twin thermal power units sharing one dry cooling tower, power production is easily impacted by variations in cooling tower performance. Circulating cooling water pre-mixing (CCWPM) was proposed to maximize cooling potential of the tower under varying working conditions, thereby enhancing power generation for the whole power plant. An integrated simulation model was established to capture cooling behaviors of the shared tower and operating characteristics of both power units before and after implementing CCWPM. Simulation results indicated that CCWPM effectively mitigates the uneven distribution of airflow temperature inside the shared tower and reduces the high-temperature area. As power load of one specified unit drops, both units experience a decline in power cycle efficiency, and the implementation of CCWPM further decreases the efficiency of this unit while increasing the efficiency of the adjacent unit. The whole power plant benefits from CCWPM in power generation, particularly when the unit with radiators directly facing the crosswind operates at a low power load. Power generation efficiency achieves an absolute increase of 0.43 % via CCWPM for the upwind unit and its adjacent unit operating at 40 % and 100 % power loads and the shared tower working at 8 m/s crosswind speed.

Novel meta-learning for few-shot bearing fault diagnosis under varying working conditions

In practical engineering, large amount data and variable working conditions poses a challenge to most existing Deep Learning(DL) methods. To solve this problem, this paper proposes a new meta-learning approach. Under the condition of limited data, the fault diagnosis under variable working conditions is regarded as a problem with fewer lenses, and the fault diagnosis of few samples across working scenes is carried out based on the Model-Agnostic Meta-Learning(MAML). Gradient-by-gradient rules are used for parameter optimization to achieve an efficient representation of these tasks. Then, the attention mechanism is applied to improve the efficiency of the training. Finally, experiments verified the fault diagnosis accuracy under various working conditions.

Multiscale Conditional Adversarial Networks based domain-adaptive method for rotating machinery fault diagnosis under variable working conditions

Deep learning has been increasingly used in health management and maintenance decision-making for rotating machinery. However, some challenges must be addressed to make this technology more effective. For example, the collected data is assumed to follow the same feature distribution, and sufficient labeled training data are available. Unfortunately, domain shifts occur inevitably in real-world scenarios due to different working conditions, and acquiring sufficient labeled samples is time-consuming and expensive in complex environments. This study proposes a novel domain adaptive framework called deep Multiscale Conditional Adversarial Networks (MCAN) for machinery fault diagnosis to address these shortcomings. The MCAN model comprises two key components. Constructed by a novel multiscale module with an attention mechanism, the first component is a shared feature generator that captures rich features at different internal perceptual scales, and the attention mechanism determines the weights assigned to each scale, enhancing the model’s dynamic adjustment and self-adaptation capabilities. The second component consists of two domain classifiers based on Bidirectional Long Short-Term Memory (BiLSTM) leveraging spatiotemporal features at various levels to achieve domain adaptation in the output space. The deep domain classifier also captures the cross-covariance dependencies between feature representations and classifier predictions, thereby improving the predictions’ discriminability. The proposed method has been evaluated using two publicly available fault diagnosis datasets and one condition monitoring experiment. The results of cross-domain transfer tasks demonstrated that the proposed method outperformed several state-of-the-art methods in terms of transferability and stability. This result is a significant step forward in deep learning for health management and maintenance decision-making for rotating machinery, and it has the potential to revolutionize its future application.

Multi-source domain adaptation using diffusion denoising for bearing fault diagnosis under variable working conditions

Transfer learning of multi-source domain adaptation seems a promising way for fault diagnosis of roller element bearings under variable working conditions. Data imbalance affects the performance of multi-source domain adaptation greatly and is expected to be solved by GAN. However, GAN-based transfer learning diagnosis models suffer pattern collapse and training instability, leading to unsatisfying diagnosis results in practical engineering. This paper proposes a denoising diffusion multi-source domain adaptation model (DDMDA). The proposed model uses diffusion denoising, which has better performance and is simpler to train than GAN, to generate shifted source domains for solving the data imbalance problem. A new noise prediction structure in diffusion denoising named Utrans-net, is constructed to restore the data distribution in the shifted source domain. Also, a multiple-domain discriminator structure is designed to extract features from multiple source domains to solve the issue of variable working conditions. Advanced models are used in this paper to compare with the proposed model for validation. Experimental demonstrations show that the proposed model is superior to the comparison models with satisfying performance.

A multi-objective co-optimization method of controller parameters for the overall system of small pressurized water reactor

Small pressurized water reactors (SPWRs) normally have complicated and varying working conditions, and become an important direction for nuclear energy development. The overall system of SPWR, including the reactor coolant average temperature control system, the feedwater control system of the once-through steam generator (OTSG), the steam turbine speed control system and the steam dump control system, is used to guarantee the safe and stable operation. The overall system of SPWR is designed with multiple PI controllers. To accomplish intelligent self-tuning of PI controller parameters, reduce the dependence on human engineering experience, and improve the control performance of the overall system, a multi-objective co-optimization method(MMOCO) with PI controller parameters for overall system of SPWR is proposed. Firstly, based on the coordinated control theory and PI control algorithm, the overall system of SPWR is established. Then, a MMOCO for the overall system of SPWR is designed with the Non-Dominated Sorting Genetic Algorithm-II (NSGA-II). Finally, to compare the performance of the controller parameters derived from the MMOCO and the controller parameters obtained via the engineering tuning method, step load decrease transients are selected. The results show that the ability to control the overall system of SPWR is effectively improved after applying the MMOCO.

Design and workspace analysis of a cable-driven space capture robot for noncooperative targets

Capturing noncooperative targets in space has garnered continuous research interest in aerospace applications. This study addresses the demands of large-scale, multifaceted activities and varied working conditions for space capture missions by designing a space capture robot composed of multiple cable-driven manipulators operating in parallel. First, single- and multi-segment cable-driven robot models were designed, and a geometric model was subsequently built. The optimal number of segments was determined by analysing the condition number of a Jacobian matrix using the Monte Carlo method. Subsequently, based on the constant-curvature assumption, a kinematic model of the cable-driven space capture robot was formulated, and capture methods for different capture targets were designed using the Monte Carlo method. Finally, an eight-segment cable-driven robot prototype was developed, and compliance and driving experiments were conducted. This robot exhibits promising application potential for space noncooperative target capture and can be feasibly manufactured using on-orbit 3D machining technology.

Rolling Bearing Fault Diagnosis in Agricultural Machinery Based on Multi-Source Locally Adaptive Graph Convolution

Maintaining agricultural machinery is crucial for efficient mechanized farming. Specifically, diagnosing faults in rolling bearings, which are essential rotating components, is of significant importance. Domain-adaptive technology often addresses the challenge of limited labeled data from a single source domain. However, information transfer can sometimes fall short in providing adequate relevant details for supporting target diagnosis tasks, leading to poor recognition performance. This paper introduces a novel fault diagnosis model based on a multi-source locally adaptive graph convolution network to diagnose rolling bearing faults in agricultural machinery. The model initially employs an overlapping sampling method to enhance sample data. Recognizing that two-dimensional time–frequency signals possess richer spatial characteristics in neural networks, wavelet transform is used to convert time series samples into time–frequency graph samples before feeding them into the feature network. This approach constructs a sample data pair from both source and target domains. Furthermore, a feature extraction network is developed by integrating the strengths of deep residual networks and graph convolutional networks, enabling the model to better learn invariant features across domains. The locally adaptive method aids the model in more effectively aligning features from the source and target domains. The model incorporates a Softmax layer as the bearing state classifier, which is set up after the graph convolutional network layer, and outputs bearing state recognition results upon reaching a set number of iterations. The proposed method’s effectiveness was validated using a bearing dataset from Jiangnan University. For three different groups of bearing fault diagnosis tasks under varying working conditions, the proposed method achieved recognition accuracies above 99%, with an improvement of 0.30%-4.33% compared to single-source domain diagnosis models. Comparative results indicate that the proposed method can effectively identify bearing states even without target domain labels, showcasing its practical engineering application value.

Cross-domain fault diagnosis network based on attributes and features transfer with dual classifier under limited and unbalanced datasets

Abstract Deep learning (DL)-based methods have demonstrated significant success in fault diagnosis owing to their robust feature extraction and non-linear fitting capabilities. Meanwhile, their remarkable performance is accompanied by constant operating conditions and sufficient monitoring data. However, in real engineering environments, variable working conditions or limited and unbalanced data are common, which can widen the gap between fault diagnosis methods and real industrial applications. In this paper, we proposed a cross-domain fault diagnosis network based on a dual classifier (CFDNet) with input being limited and unbalanced data to learn attributes and features for unsupervised domain adaptation. We found that the diagnostic performance is commonly bounded by the underlying knowledge, especially feature extraction from original data. Therefore, we designed a new feature encoder with features and relationships, i.e. using a convolutional neural network and graph convolutional network, which improves extraction efficiency while retaining valuable information. Then, we discovered that enforced feature transfer can lead to negative transfer. To mitigate this, we present a feature and attribute transfer framework, which not only achieves features transfer but also enables attributes transfer. Furthermore, it was noted that limited and unbalanced datasets can introduce label bias and lead to biased model training. Hence, we designed dual classifiers to improve the probability of high-confidence final prediction by synthesizing diagnostic results. Comprehensive experiments conducted on three case studies demonstrate the effectiveness and superiority of our method for cross-domain fault diagnosis under limited and unbalanced datasets, which outperforms state-of-the-art methods in this study.

A meta-learning method for smart manufacturing: Tool wear prediction using hybrid information under various operating conditions

Accurate tool wear prediction during machining is crucial to manufacturing since it will significantly influence tool life, machining efficiency, and workpiece quality. Although existing data-driven methods can achieve competitive performance in tool wear prediction, their main emphasis is on fixed operating conditions with sufficient training samples, which is impractical in engineering practice. This implies that predicting tool wear values under variable working conditions with insufficient data is still a challenge owing to the difference in data distributions in complex tool wear mechanisms. Besides, having no access to samples in new conditions is another challenge for tool wear prediction in engineering practice. To address these issues, we develop a hybrid information model-agnostic domain generalization (H-MADG) method to provide appropriate initial model parameters that can be fast adaptative to the new conditions after fine-tuning. Additionally, we construct hybrid information as model input by fusing process information with temporal properties derived by neural networks, and the hybrid information can offer more useful prior knowledge about the machining process. Experimental results on NASA milling data show that compared with contrastive techniques, the RMSE of the proposed H-MADG method is reduced by an average of 36.81 %, which can achieve a low average RMSE value of 0.0904 with 15 cases under five different network architectures. We also investigate several crucial impact factors of the H-MADG method and summarize corresponding analysis and suggestions.

An adjustable stiffness vibration isolator implemented by a semicircular ring

Nonlinear vibration isolators with quasi-zero-stiffness characteristics offer broadband vibration isolations, while passive quasi-zero-stiffness isolators are unable to adapt to variable working conditions. To address the issue, a nonlinear semi-active vibration isolator based on a semicircular ring structure is proposed. As an elastic component, the semicircular ring exhibits nonlinear stiffness under a vertical compression at the centre. With a semi-active control of the horizontal distance between the two ends of the semicircular ring, a pre-deformed state is induced, and the stiffness characteristics are adjusted accordingly. A chained beam-constraint model is adopted to characterize the pre-deformed process and the vertical compression process, respectively. The adjustable nonlinear restoring force is accurately calculated and adopted for vibration analysis. A harmonic balance method is used to analyze the frequency responses of the isolator under different working conditions. The semicircular ring exhibits softening stiffness under a vertical compression, and the quasi-zero-stiffness region can be effectively adjusted by the pre-deformed process. With the vertical compression reaching a threshold, negative stiffness is presented along with a snap-through phenomenon due to the bi-stability. With the semi-active pre-deformed, the isolator demonstrates different resonant frequencies and different risks in the occurrence of the snap-through during vibration. To avoid the snap-through-induced impact, the control threshold of the pre-deformed is determined, depending on the payload mass, the excitation amplitude, and the damping coefficient. Experiments are performed to validate the adjustable stiffness and the frequency responses of the isolator, and both the jump phenomenon and the snap-through phenomenon are observed.

A novel transformer-based few-shot learning method for intelligent fault diagnosis with noisy labels under varying working conditions

Recent years have witnessed the success of Few-shot Learning (FSL) methods in equipment reliability enhancement and fault diagnosis, by virtue of learning from limited data and adapting to new operating conditions. However, due to sensor bias, manual collection, and mislabeling, label noise is inevitably introduced into the dataset, which further reduces the quality of supervised information contained in the few-shot dataset, posing significant challenges for accurate fault diagnosis. In this paper, the problem of Few-shot Fault Diagnosis with Noisy Labels (FFDNL) is studied for the first time, and a novel method named Enhanced Transformer with Asymmetric Loss Function (ETALF) is proposed. ETALF leverages the self-attention mechanism of the transformer to dynamically measure the similarity between fault samples in the support set to enhance the model's robustness against label noise, then naturally aggregates the similar samples into corresponding correct prototypes. Furthermore, an asymmetric loss function is designed, which adaptively assigns the model with larger penalties for incorrect category predictions and smaller penalties for correct category predictions, thereby enhancing fault diagnostic performance through inherent asymmetry. Comprehensive experiments are conducted on two benchmark datasets, and the compared results with representative approaches validate the effectiveness of our proposed ETALF in performing intelligent fault diagnosis using limited and noise-labeled data under varying working conditions, which achieves accuracies of 97.77% and 95.78% with 0.2 noisy-level labels during meta-training and meta-testing on the CWRU and KAIST datasets, respectively.

Modelling of electromechanical coupling dynamics for high-speed EHT system used in HEV and characteristics analysis

As the development of hybrid electric vehicle (HEV) to high-speed, heavy-load, the electromechanical coupling vibration becomes the bottleneck in high-speed electromechanical hybrid transmission (EHT) system. To explore the dynamics characteristics and optimization direction for high-speed EHT system, an electromechanical coupling dynamics model under multi-source excitations is established with lumped-distributed parameter method and verified with experiment data. The dynamics model shows higher accuracy in both time and frequency domain analysis. On this basis, firstly, the comparison between lumped-shaft and distributed-shaft model is studied under time and frequency domain. Comparing with the lumped-shaft model, the distributed-shaft model shows higher accuracy, and can better reflect the coupling vibration of multi-stage planetary gears (PGs). Secondly, the inherent vibration model is derived, and the effect of electromechanical coupling and high-speed working conditions on inherent vibration characteristics are studied. Thirdly, a new method called ‘machine-electricity-magnet coupling interface’ is proposed to reveal the coupling vibration phenomenon. In addition, the signal of stator current and electromagnetic torque includes the frequency of PGs, which is a basis on the fault diagnosis and state monitor of EHT system. Last but not the least, the vibration acceleration of EHT system is analysed under variable speed and load working conditions. Rotation speed and gear meshing force is found as the main influencing factor of series EHT system, and the specific optimization direction is also given.

Multi-condition adaptive detail characterization model of magnetorheological dampers and experimental verification

The theoretical model for predicting the damping characteristics of magnetorheological dampers (MRDs) is significant for enhancing the design efficiency of the control algorithm. However, some existing theoretical models face limitations in characterizing MRD damping characteristics simultaneously in terms of nonlinear detail characterization and adaptability to variable working conditions. Therefore, this paper proposed the Composite Double-Boltzmann (CDB) model combining the Double-Boltzmann (DB) function widely used in the field of biology and chemistry for its strong nonlinear characterization capability. Utilizing this model to fit the sinusoidal vibration testing data of the MRD prototype under variable combination working conditions, obtaining quantitative relationships between the undetermined parameters in the CDB model and the excitation current, vibration frequency, and amplitude to enable the model to address both the nonlinear details characterization of MRDs and adaptability to variable working conditions. Subsequently, the validity of the quantitative relationships were verified by comparing the calculated parameter values using the quantitative relationships with the original accurate parameter values. In order to verify the validity of the CDB model, extensive unknown working condition vibration tests were conducted on the MRD prototype under variable excitation currents, vibration frequencies, amplitudes and random excitation working conditions, employing the CDB and Tanh models to predict the damping characteristics, to compare to demonstrate the CDB model’s capability of adapting to variable working conditions while accurately characterizing the nonlinear details of MRD damping characteristics.

A two-stage importance-aware subgraph convolutional network based on multi-source sensors for cross-domain fault diagnosis

Graph convolutional networks (GCNs) as the emerging neural networks have shown great success in Prognostics and Health Management because they can not only extract node features but can also mine relationship between nodes in the graph data. However, the most existing GCNs-based methods are still limited by graph quality, variable working conditions, and limited data, making them difficult to obtain remarkable performance. Therefore, it is proposed in this paper a two stage importance-aware subgraph convolutional network based on multi-source sensors named I2SGCN to address the above-mentioned limitations. In the real-world scenarios, it is found that the diagnostic performance of the most existing GCNs is commonly bounded by the graph quality because it is hard to get high quality through a single sensor. Therefore, we leveraged multi-source sensors to construct graphs that contain more fault-based information of mechanical equipment. Then, we discovered that unsupervised domain adaptation (UDA) methods only use single stage to achieve cross-domain fault diagnosis and ignore more refined feature extraction, which can make the representations contained in the features inadequate. Hence, it is proposed the two-stage fault diagnosis in the whole framework to achieve UDA. In the first stage, the multiple-instance learning is adopted to obtain the importance factor of each sensor towards preliminary fault diagnosis. In the second stage, it is proposed I2SGCN to achieve refined cross-domain fault diagnosis. Moreover, we observed that deficient and limited data may cause label bias and biased training, leading to reduced generalization capacity of the proposed method. Therefore, we constructed the feature-based graph and importance-based graph to jointly mine more effective relationship and then presented a subgraph learning strategy, which not only enriches sufficient and complementary features but also regularizes the training. Comprehensive experiments conducted on four case studies demonstrate the effectiveness and superiority of the proposed method for cross-domain fault diagnosis, which outperforms the state-of-the art methods.

Few-shot bearing fault diagnosis by semi-supervised meta-learning with graph convolutional neural network under variable working conditions

Aiming at the problems of low accuracy and weak generalization ability in fault diagnosis caused by complex working conditions and limited fault samples of bearings, a few-shot bearing fault diagnosis method by semi-supervised meta-learning with simplifying graph convolutional neural network (semi-meta-sgc) is proposed. Firstly, the time domain signal is converted into the frequency domain by fast Fourier transform (FFT). Secondly, considering each spectrum sample as a node, a k-nearest neighbor (KNN) graph is constructed according to the adjacencies between nodes, thus transforming the bearing fault classification into a graphical node classification. Then, relying on meta-learning, graph convolutional neural network, and semi-supervised learning, a high-accuracy node classifier is obtained by using a small number of training samples to achieve the classification of bearing faults under variable working conditions. Finally, the proposed method is verified by four cases and compared with other methods. The results show that this method has higher recognition accuracy and generalization performance.

Effect of variable conditions on transient flow in a solid–liquid two-phase centrifugal pump

To investigate the influence of altering operational parameters on the transient flow characteristics within the flow channel of a solid–liquid two-phase centrifugal pump, this study employs a particle model based on the Euler–Euler method. Utilizing the standard k–e model, flow field simulations are conducted using the ANSYS-CFX software. Specifically, the study focuses on throttle regulation scenarios, monitoring and comparing the external and internal flow parameters of the solid–liquid two-phase pump with those of a clear water medium centrifugal pump. The results indicate notable modifications in head, efficiency, and shaft power due to the presence of solid particles in the two-phase flow. Decreasing flow rates during throttle regulation lead to fluctuations in pressure and turbulence energy distribution. Furthermore, under identical operational conditions, the variable working conditions of the solid–liquid pump result in increased flow rates of solid-phase particles near the impeller's outer edge, with particles shifting toward the middle and tail of the vane suction surface. This phenomenon exacerbates wear on the vane tail of the suction surface. Moreover, the study identifies that changing operational conditions in the solid–liquid dual-flow centrifugal pump contribute to increased axial forces, consequently leading to pump vibrations. Overall, this research elucidates the transient flow characteristics within the flow channel of solid–liquid two-phase centrifugal pumps under varied operational conditions, serving as a foundational reference for assessing the stability of such pumps.

Analysis on axial force characteristics of variable valve and variable speed adjustment of centrifugal pump

In order to study the evolution mechanism of axial force characteristics of centrifugal pump in the transient process of variable working conditions, the IS80-65-160 centrifugal pump is taken as the research object, and four flow operating points of centrifugal pump 1.0Qr, 0.8Qr, 0.6Qr, and 0.4Qr are selected. Non-constant numerical simulation is carried out, and on the basis that the numerical calculation results of the external characteristics are basically consistent with the experimental results, the changes of the axial force of the impeller during the flow reduction process of the variable valve adjustment and the variable speed adjustment of the centrifugal pump are studied, respectively. The results show that as the flow rate decreases, the force of the variable valve regulating impeller cover plate increases, and the force of the variable speed regulating cover plate decreases. Under the two adjustment methods, the axial force on the pressure surface of the blade becomes smaller, the axial force on the suction surface of the blade is basically unchanged, and the pulsation coefficient of the impeller axial force increases first and then decreases. The axial force pulsation coefficient reaches the maximum value in the variable valve adjustment condition of 0.8Qr and the variable speed adjustment condition of 0.6Qr, respectively, and the axial force on the outer wall of impeller cover plate has the greatest influence on the main frequency axial force ripple coefficient; the axial force acting on the outer wall surface of the front cover plate of the impeller and the inner wall surface of the front and rear cover plates increases with the increase in the radius, The axial force acting on the outer wall surface of the rear cover plate of the impeller has axisymmetrical properties. This study revealed the evolution mechanism of the axial force characteristics of the centrifugal pump impeller during the transient process of variable working conditions, and the research results can provide a reference for improving the stable operation of the centrifugal pump.

Rolling Bearing Fault Diagnosis Based on MResNet-LSTM

In response to the challenges of weak fault characteristics and complex and variable working conditions of roll-ing bearings in strong noise environments, a diagnostic methodology is proposed. This method integrates a multi-scale residual neural network (MResNet) and a long short-term memory (LSTM) neural network to achieve fault diagnosis of bearings under both constant and variable rotational speed working conditions. First, complete an ensemble empirical mode decomposition with adaptive noise (CEEMDAN) that is applied for vibration signal denoising. Secondly, a dropout layer is introduced between multi-scale residual blocks to prevent net-work overfitting, and the powerful time series information capture ability of LSTM is combined to improve di-agnostic accuracy. Finally, experimental verification is conducted using constant and variable speed bearing data sets. The results show that the proposed approach maintains strong diagnostic capabilities when the rotat-ing speed conditions change, and the signal is heavily polluted by noise.

Adaptive angle-weighted cumulative sum for interpretable machine condition monitoring

Health indicators (HIs) are important indicators in machine condition monitoring (CM) to determine whether downtime is required for maintenance. It can quantitatively characterize the degradation stages of machines in different operating conditions and provide theoretical guidance for predictive maintenance of machines. Therefore, the accurate construction of HIs is of great significance for the safe operation and economic savings of machines. In recent years, extensive research has been conducted on numerous deep learning-based methods for HI construction. However, these approaches heavily depend on historical degradation data from machines. Furthermore, the HIs constructed through deep learning often lack sufficient interpretability, thus restricting their widespread application. To tackle this challenge, an innovative HI construction method tailored for interpretable machines is presented, employing the adaptive angle-weighted cumulative sum (AAWCS) approach derived from a thorough investigation of machine failure degradation mechanisms. Firstly, low-pass filtering is employed to mitigate the effects of rotational speed fluctuations and background noise interference existing in machines for actual industrial applications. Besides, the arctangent transform is applied to improve the signal-to-noise ratio of the original run-to-failure (RF) vibration signal collected by the machine, which is transformed into the angle-domain for the next step of analysis. On this basis, an adaptive angle weight (AAW) assignment mechanism is performed to obtain an angle-domain weight matrix related to the degree of machine failure deterioration in depth, which has a clear interpretability. Besides, the Shannon entropy is employed to quantitatively evaluate the angle-domain weight matrix for measuring correlations with the degree of machine failure deterioration, which enables the accurate construction of HI that can characterize the machine health status. Finally, the validation of the AAWCS method is performed with a typical bearing RF data and two real industrial machines RF data, which demonstrate that the AAWCS method can be generalized to adaptive CM of wind turbines under variable working conditions.

A novel adaptive trajectory tracking control for complex environments based on accelerated back-propagation neural network

This work addresses the issue of trajectory tracking accuracy degradation for vehicle dynamics parameters and road conditions changing, an adaptive trajectory tracking controller is designed for the intelligent vehicle. Based on the yaw dynamics model, a model predictive controller (MPC) is developed for the trajectory tracking process. Meanwhile, the impacts of different control parameters on trajectory tracking at various speeds are compared. And the change rule between optimal control parameter values and reference speeds is determined by cubic polynomial fitting to ensure minimal error during the trajectory following process. Then, an estimator for vehicle tire cornering stiffness and road adhesion coefficient is designed and tested based on the accelerated back-propagation neural network (ABPNN) model, with its estimation values being compared to those obtained through recursive least square method. Last, a co-simulation platform combining MATLAB/Simulink and CarSim software is established to conduct the trajectory tracking experiment under variable working conditions. Experimental results show that the proposed adaptive controller not only exhibits high accuracy in trajectory tracking but also demonstrates excellent stability and adaptability under medium to low speed conditions, particularly when encountering changing road conditions.