Electronic function test system for high-voltage PTC electric heater

Abstract

The application discloses an electronic function test system of a high-voltage PTC electric heater, relates to the technical field of PTC electric heaters, and collects extreme temperature, humidity and voltage disturbance data in a programmable multi-dimensional environment simulation cabin, predicts temperature deviation of the high-voltage PTC electric heater by using a multi-channel fusion and compensation model, adjusts output power in real time by feedforward and feedback in an adaptive controller, and finally optimizes the model and gain parameters in multiple scenes through dynamic calibration and closed-loop verification, so that a complete system which can cope with large environmental changes and has high temperature control precision is formed, the stability and safety of the high-voltage PTC electric heater under severe working conditions are significantly improved, overshoot and lag are reduced, energy utilization efficiency is improved, the system is suitable for scenes such as industrial heating and vehicle heating which require high reliability and fast response, and maintenance cost can be effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of PTC electric heater technology, specifically to an electronic function testing system for high-voltage PTC electric heaters. Background Technology

[0002] In modern industrial production and vehicle heating systems, high-voltage PTC (Positive Temperature Coefficient) electric heaters are widely used due to their rapid heating, high thermal efficiency, and self-limiting current characteristics. However, these heaters often face more stringent challenges in extreme and variable external environments or complex operating conditions. For example, during the low-temperature start-up phase in cold regions during winter, ambient temperatures can plummet to tens of degrees below zero Celsius, accompanied by high humidity or snow cover. Similarly, in remote construction areas or areas with temporary power grids, grid voltage often fluctuates drastically over a wide range, even experiencing momentary power outages or overvoltages. Under these multiple environmental disturbances, traditional constant indoor environmental calibration methods cannot accurately reflect actual operating conditions, leading to risks such as temperature overshoot, hysteresis, or failure to achieve the expected heating effect after PTC electric heaters are put into field use. To ensure the stability and safety of these critical components in extreme environments, it is necessary to employ a test platform with programmable temperature, humidity, and voltage simulation capabilities during the design and verification process, and to collect detailed data using a multi-channel sensor network to accurately evaluate their operating characteristics under harsh conditions.

[0003] However, existing industrial testing systems generally lack the ability to globally simulate and dynamically control extreme external environments, making it difficult to systematically monitor and calibrate the temperature control performance of PTC electric heaters under real-world scenarios involving rapid temperature changes, high humidity shocks, and significant fluctuations in grid voltage. In particular, when ambient temperature or humidity and supply voltage change significantly within a short period, traditional temperature control strategies using fixed parameters or simple PID control often fail to promptly detect and quantify the impact of external factors on the resistive characteristics of the PTC element. This results in a significant deviation between the actual heating power and the temperature target, leading to problems such as slow power response, temperature overshoot, or under-adjustment. This not only reduces energy efficiency but may also pose safety hazards to the equipment. Therefore, there is an urgent need for a complete technical solution capable of multi-channel data fusion and adaptive temperature control in non-constant environments. This solution would provide extreme condition testing through a programmable simulation chamber, accurately predict temperature deviations using an environmental compensation model, and introduce a closed-loop adaptive control method to dynamically adjust the heater power, thereby ensuring that the high-voltage PTC electric heater maintains a stable, accurate, and efficient temperature output under various complex operating conditions.

[0004] Therefore, the present invention provides an electronic function testing system for high-voltage PTC electric heaters. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides an electronic functional testing system for high-voltage PTC electric heaters. By collecting extreme temperature, humidity, and voltage disturbance data in a programmable multi-dimensional environmental simulation chamber, and using a multi-channel fusion and compensation model to predict the temperature deviation of the high-voltage PTC electric heater, the system adjusts the output power in real-time through feedforward and feedback in an adaptive controller. Finally, through dynamic calibration and closed-loop verification, the model and gain parameters are optimized in multiple scenarios, forming a complete system capable of handling significant environmental changes and possessing high temperature control accuracy. This significantly improves the stability and safety of high-voltage PTC electric heaters under harsh operating conditions, thus solving the technical solutions described in the background art.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solution: an electronic function testing system for a high-voltage PTC electric heater, comprising: when it is detected that the high-voltage PTC electric heater needs to simulate an extreme environment, a multi-dimensional environment simulation chamber performs programmable temperature and humidity control and variable voltage source operation, periodically adjusts and collects environmental parameters in real time, and generates multi-source basic data containing extreme disturbances;

[0009] After the data is collected, wavelet filtering and feature extraction algorithms are used on the environmental parameter data to generate denoised and normalized signals and construct an environmental compensation model to output the environmental disturbance compensation amount.

[0010] When the environmental compensation model outputs the environmental disturbance compensation amount in real time and detects that the temperature deviation is greater than expected, the adaptive temperature control algorithm combines the feedforward and feedback paths, and superimposes the corresponding disturbance information in the online adjustment of PID gain to form a corrected power control signal.

[0011] When new environmental disturbance data appears, multi-scenario tests are performed in the environmental simulation chamber and the temperature response curve and environmental disturbance compensation amount are recorded. The model weights and control gain parameters are optimized synchronously based on the cumulative temperature deviation evaluation results.

[0012] Furthermore, programmable heat and cold sources, adjustable humidity generators, and variable voltage sources are installed inside the simulation chamber. By selecting composite materials resistant to temperature difference shock, heat insulation layers and isolation layers are laid out inside, and an air duct system is added.

[0013] Inside the simulation chamber, a sensor array is constructed according to high temperature zone, low temperature zone, and strong convection zone to collect environmental parameter components and construct an overall environmental value. When the overall environmental value rises rapidly, an alert is sent to the outside.

[0014] Furthermore, all original signals are first time-base aligned, and an adaptive filtering strategy based on wavelet transform is adopted to decompose each channel signal at different scales. After filtering in different frequency bands, the signals are reconstructed, key features are extracted and recorded as multi-channel feature vectors, and multi-channel signals are obtained.

[0015] Furthermore, a high-dimensional input vector is formed by integrating the multi-channel signals, the components of the multi-channel feature vectors, and the environmental comprehensive value.

[0016] An environmental compensation model is constructed based on a high-dimensional input vector, outputting the compensation amount for environmental disturbances caused by temperature deviation, and an online learning method is used to iteratively update the weight vector of the environmental compensation model.

[0017] Furthermore, the control signal of the adaptive temperature controller is divided into a feedforward compensation component and a feedback component. The environmental disturbance compensation amount is multiplied by the gain coefficient after passing through a nonlinear mapping operator to obtain the feedforward signal.

[0018] The conventional PID or incremental PID controller is transformed into an adaptive PID controller, so that its gain coefficient can be gradually corrected during operation, and the environmental disturbance compensation is incorporated into the gain update rule.

[0019] Furthermore, the feedforward compensation component and feedback component are calculated based on the feedforward algorithm and feedback algorithm to obtain the total output. The total output is then converted into the actual heater power or duty cycle adjustment signal to control the output heat of the high-voltage PTC electric heater.

[0020] The environmental disturbance compensation amount is updated based on the latest environmental data or model predictions. The adaptive PID controller then combines the current error and the environmental disturbance compensation amount to complete the adaptive adjustment of the gain.

[0021] Furthermore, after configuring several sets of extreme test scenarios, the environmental parameters and control signals are recorded synchronously, and the temperature response curves, temperature control signals and various environmental parameter variables under several scenarios are output.

[0022] A cumulative temperature deviation functional is constructed based on the environmental compensation model and its predictive effect on environmental disturbance compensation.

[0023] If the cumulative temperature deviation functional is higher than expected, the backhaul data is used for quadratic regression or online learning updates to correct the weight vector or nonlinear mapping function of the environmental compensation model.

[0024] Furthermore, for the adaptive PID controller, the initial gain is reset or the gain update rule is adjusted based on the overshoot and settling time found in the test results.

[0025] If the modification of the environmental compensation model causes a significant change in the amount of environmental disturbance compensation, the gain coefficient or parameters of the nonlinear function of the feedforward compensation need to be updated simultaneously to obtain the adjusted adaptive temperature control strategy.

[0026] Furthermore, for the adjusted environmental compensation model-PID controller combination, a convergence index is defined to evaluate the gradual convergence of the self-learning algorithm. If the convergence index continues to oscillate or rise during long-term testing, the self-learning convergence index is modified.

[0027] Furthermore, the temperature response curves, cumulative temperature deviation functionals, and convergence indices collected during the long-term testing process are visualized and comprehensively analyzed to determine whether the expected industrial application indicators have been achieved. If the long-term verification shows that the industrial application indicators do not meet expectations, further iterative optimization can be performed based on the verification results.

[0028] (III) Beneficial Effects

[0029] This invention provides an electronic function testing system for high-voltage PTC electric heaters, which has the following beneficial effects:

[0030] A multi-dimensional environmental simulation chamber (including a programmable temperature source, humidity source, and voltage source) was constructed and a precision sensor network was deployed. The chamber's internal temperature T... env (t), Relative humidity inside the cabin H env (t), power supply voltage V env (t) and heater current I load Fine-grained acquisition is achieved on multiple channels such as (t), enabling the acquisition of real extreme environment data and the establishment of rich test scenarios in the early stage;

[0031] By cleaning and extracting features from multi-channel data and combining them with historical operation records, an environmental compensation model is constructed to output the environmental disturbance compensation amount ΔT. pred Using the denoised signal and the multi-channel feature vector F(t) obtained from wavelet decomposition and high-dimensional regression, the PTC element resistance change and temperature deviation trend can be accurately predicted. Compared with the conventional linear method, this compensation model can maintain higher accuracy in dynamic and complex external disturbances because it integrates nonlinear mapping and environmental disturbance information.

[0032] The environmental disturbance compensation amount ΔT is calculated using an adaptive temperature control algorithm. pred (t) combined with conventional feedback loop to achieve bidirectional control of feedforward and feedback, so that the control signal u(t) can correct the power output in advance before the environmental change has fully affected it. By implementing this strategy online, incremental PID or adaptive PID can automatically correct the gain in different scenarios, avoiding temperature overshoot and long-term lag. Compared with traditional constant parameter PID, the solution can still maintain fast and stable temperature control under conditions such as large fluctuations in external voltage or extreme cold and high humidity.

[0033] By setting up multiple extreme environmental scenarios, the actual temperature T was recorded. meas (t) and environmental disturbance compensation amount ΔT pred Data such as (t) and control signal u(t); then the model-controller is coordinated and optimized; and long-term operation and self-learning convergence test are performed. The overall operating quality is judged by combining the cumulative temperature deviation functional γ. This process can effectively discover the potential defects of the system under extreme conditions, and continuously improve the temperature control accuracy and stability through iterative calibration, so as to achieve high reliability and high efficiency under harsh working conditions.

[0034] In summary, this solution achieves precise temperature control and optimal energy utilization under extreme conditions through the close collaboration and data closed loop of four major modules: environmental simulation chamber, multi-channel data processing, adaptive temperature control algorithm, and dynamic calibration mechanism. It can accurately capture the disturbances of the external environment to the PTC electric heater and flexibly adjust the compensation model and control strategy. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the electronic function testing system for the high-voltage PTC electric heater of the present invention. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Please see Figure 1 This invention provides an electronic functional testing system for a high-voltage PTC electric heater, comprising:

[0038] Step 1: When it is detected that the high-voltage PTC electric heater needs to be tested for full-domain environmental disturbance, the multi-dimensional environmental simulation chamber scheduling interface is called. The programmable temperature and humidity control unit and variable voltage source are used to controllably switch the temperature, humidity and grid voltage in the simulation chamber to form various extreme and gradual working conditions, and to record at fixed points and times, so that the multi-channel sensors can output high-resolution operating data in real time.

[0039] Step one includes the following:

[0040] Step 101: Programmable structural design of the multi-dimensional environment simulation cabin

[0041] The simulation chamber is equipped with programmable heat and cold sources, an adjustable humidity generator, and a variable voltage source. By selecting composite materials that are resistant to temperature difference shocks and laying out heat insulation and isolation layers inside, the airtightness and safety of the simulation chamber under harsh working conditions are ensured. An air duct system is also added to create temperature or humidity gradients between different areas to enhance the diversity of the simulation scenarios.

[0042] When in use, the programmable heat source, cold source, humidity control unit and voltage control module are combined into one unit to form a chamber that can simultaneously apply multiple environmental disturbances, providing a complex stress scenario for subsequent data acquisition and environmental compensation, far exceeding the conventional testing capabilities of a single constant temperature or humidity chamber.

[0043] Step 102: Deployment of Precision Sensor Network and High-Resolution Data Acquisition

[0044] Multiple high-sensitivity temperature sensors were distributed in key areas such as high-temperature zone, low-temperature zone, and strong convection zone within the simulation chamber, and a temperature sensor array was constructed to collect the chamber temperature T. env The distribution of (t); a humidity sensor array is deployed corresponding to the temperature sensor to capture the relative humidity H inside the cabin in real time. env The local gradient and overall trend of change of (t);

[0045] A voltage sensor and a current sensor are installed at the power supply inlet and the heater load end, respectively, to record the power supply voltage V. env (t) and the current I flowing through the heater load Environmental parameter components such as (t);

[0046] All signals collected by the sensors are based on a unified time base, and each data record is marked with a timestamp τ (τ∈[0,+∞)) with millisecond-level or higher precision. To ensure that the evaluation of the disturbance of each channel is reasonable, each component can be transformed or normalized. Alternatively, a linear transformation or standardization operation can be introduced before summarizing the multi-channel signals. Different dimensions such as environmental parameters can be normalized to similar intervals, and then wavelet filtering or feature extraction weighted matrix scaling correction can be performed.

[0047] To preliminarily quantify the multidimensional environmental intensity, the following comprehensive environmental value E(t) is proposed to quickly characterize the severity of the cabin environment, where: the environmental state vector is defined as follows: Wherein: T env (t) represents the cabin temperature, H. env (t) represents the relative humidity inside the cabin, V env (t) represents the voltage at the heater's power supply terminal;

[0048] Define the environment gradient vector: It represents the rate of change of temperature, humidity, and voltage over time, and is used to describe the severity of environmental changes in a short period of time.

[0049] Ideally, it can be considered as the derivative in continuous time, but in practical systems, discrete sampling (period Δt) is often used, so the derivative needs to be replaced by a difference:

[0050]

[0051] If the period Δt is not constant or there is missing data, appropriate interpolation or alignment should be performed to avoid amplifying the error.

[0052] W1 is a 3×3 diagonal or sparse matrix weighted by the current state of the environment, used to reflect the importance of each environmental quantity in different test scenarios, for example:

[0053] W1=diag(α1,α2,α3),α1,α2,α3∈[0,+∞)

[0054] W2 is a 3×3 diagonal or sparse matrix weighted by the rate of change (gradient) of the environment, used to balance the contributions of the three environmental variables to the degree of dynamic intensity, for example:

[0055] W2=diag(β1,β2,β3),β1,β2,β3∈[0,+∞)

[0056] p and q are used to define the exponents of the vector norm (usually p, q ≥ 1), controlling the penalty method for different dimensional components;

[0057] κ and γ are two nonlinear amplification factors (κ>0, γ>0), which amplify the cumulative impact of environmental intensity in the final comprehensive index in an exponential manner.

[0058] Based on this, the comprehensive environmental value E(t) can be defined as:

[0059]

[0060] When the comprehensive environmental value E(t) is at a low value, it indicates that the environmental variables and their rate of change are in a relatively stable or slightly disturbed stage in the cumulative observation of [0,t].

[0061] When the comprehensive environmental value E(t) rises rapidly and reaches a value much greater than 1, it indicates that the environmental parameters have experienced extreme conditions and / or drastic changes in the past period or at the present moment, and the system needs to increase its vigilance and compensation efforts in temperature control.

[0062] In use, it achieves high-resolution, high-synchronization data acquisition of multi-dimensional environmental disturbances within the simulation chamber, providing accurate information input for multi-channel data fusion, and through environmental comprehensive values... E (t The definition of ) can facilitate the rapid determination of extreme environmental conditions in subsequent algorithms, thereby improving the sensitivity of temperature control algorithms to sudden environmental fluctuations.

[0063] Step 2: After the multi-channel raw data sources have been synchronously collected, the data acquisition unit performs noise reduction and normalization operations on the multi-channel signal input based on wavelet filtering and feature vector extraction, and obtains the environmental disturbance compensation amount ΔT according to the nonlinear mapping operator in the environmental compensation model M. pred (t), so as to output the feedforward correction under extreme external disturbances;

[0064] Step two includes the following:

[0065] Step 201: Multi-channel data cleaning and feature extraction

[0066] First, all raw signals are time-base aligned to ensure the cabin temperature T. env (t), Relative humidity inside the cabin H env (t), power supply voltage V env (t) and heater current I load (t) are comparable at the same timestamp τ;

[0067] To filter high-frequency noise or harmonic interference while preserving abrupt changes crucial for subsequent analysis, an adaptive filtering strategy based on wavelet transform is employed. This strategy decomposes each channel signal at different scales, where: Let If we use wavelet operators, then for example, the temperature signal can be represented as:

[0068]

[0069] In the formula: φ(·) is the selected wavelet basis function (Daubechies or Meyer basis functions can be selected according to different needs);

[0070] s represents the scaling factor (s∈(0,+∞)), which captures different frequency components through multiscale analysis;

[0071] The results can be further reconstructed after filtering in different frequency bands to eliminate random high-frequency noise or low-frequency drift. Similar operations are also applicable to the cabin relative humidity H. env (t), power supply voltage V env (t) and heater current I load (t);

[0072] After the above wavelet decomposition and reconstruction, key features are extracted from the multi-channel signal and uniformly denoted as multi-channel feature vector F(t), which includes several components such as temperature change coefficient, humidity rapid fluctuation coefficient, voltage disturbance intensity and current load distribution.

[0073] Temperature change coefficient: reflects the cabin temperature T env (t) Approximates the extreme value of the first derivative at a specific time interval, characterizing the rate of transient temperature change inside the cabin; Humidity rapid fluctuation coefficient: measures the relative humidity H inside the cabin. env (t) Energy peak value in a specific frequency band of the wavelet, used to identify high humidity disturbances; Voltage disturbance intensity: relative to the supply terminal voltage V env (t) Perform short-time energy assessment to measure the transient impact of grid fluctuations on the load;

[0074] Current load distribution: from heater current I load The average load and fluctuation amplitude are extracted from (t) to determine the power change of the PTC heater.

[0075] Finally: The output of this step includes: the cleaned and reconstructed multi-channel signal denoted as...

[0076] In use, wavelet decomposition and reconstruction can preserve key abrupt changes in temperature, humidity, voltage and current at multiple scales, while effectively eliminating random noise. This ensures that subsequent model construction is based on data with a high signal-to-noise ratio. The generated feature vectors cover the instantaneous level and dynamic rate of change of the environmental state, enabling the subsequent compensation model to accurately capture the potential impact of the external environment on the PTC heater.

[0077] Step 202: Construction and Output of the Environmental Compensation Model

[0078] Integrating multi-channel signals The components of the multi-channel feature vector F(t) and the environmental comprehensive value E(t) are combined to form a high-dimensional input vector X(t), which is then integrated in the following way:

[0079]

[0080] Here, the multi-channel feature vector F(t) can contain several sub-components (such as the temperature change coefficient, the humidity rapid fluctuation coefficient, etc.), which constitute the remaining dimension of the vector;

[0081] An environmental compensation model M is constructed based on a high-dimensional input vector X(t), and the output is an estimate and compensation amount for the temperature deviation, denoted as the environmental disturbance compensation amount ΔT. pred (t), at this point, the environmental compensation model M can be set as a high-order kernel regression or deep network function, or a custom high-order polynomial form can be used, as shown in the following example:

[0082]

[0083] Where: φ(·) is a set of nonlinear mapping operators (which can be regarded as high-dimensional kernel functions or hidden layer activations of deep networks) that map the high-dimensional input vector X(t) to a higher-dimensional feature space; w is the weight vector to be trained or calibrated; ΔT pred (t) is the real-time prediction of temperature control deviation caused by environmental disturbances.

[0084] An online learning method is used to iteratively update the weight vector w of the environmental compensation model M. When there is the latest measured temperature deviation (by comparing it with the actual measured temperature) or more extreme environmental data, the weight vector w is incrementally adjusted so that the compensation model can maintain its sensitivity to extreme environments and new operating conditions during long-term operation.

[0085] In practice, combining wavelet-filtered multi-channel data, the comprehensive environmental value E(t), and higher-order mappings can quickly capture the impact patterns of external disturbances on PTC heaters and output environmental disturbance compensation in a timely manner, significantly improving adaptability to complex environments. Through dynamic correction and adaptive learning, the model's weight vector w continuously approaches the optimal value under actual operating conditions, overcoming the lag problem of static parameter PID or ordinary linear models in the face of sudden environmental changes.

[0086] The environmental disturbance compensation amount ΔT calculated by this model under different environmental scenarios (such as extremely low temperature, extremely high humidity, and strong power grid fluctuations) is pred (t) will be linked with the real-time temperature control loop to correct the setting of heating power or the target parameters of the controller, thereby ensuring that high temperature control accuracy can still be maintained under extreme external disturbances.

[0087] Step 3: When the environmental compensation model outputs the environmental disturbance compensation amount ΔT pred (t) and the actual temperature T meas When there is a large deviation in (t), the adaptive temperature control algorithm superimposes the feedforward compensation component in the feedforward compensation path and dynamically updates the gain parameter in the feedback control loop, so that the control signal u(t) corrects the power output in advance and quickly converges the temperature error.

[0088] Step three includes the following:

[0089] Step 301: Structural Design and Environmental Disturbance Compensation Integration of Adaptive Temperature Control Algorithm

[0090] Environmental disturbance compensation amount ΔT pred (t) is considered as an estimate of the potential temperature deviation caused by external factors (such as low temperature, high humidity or power grid instability), and is incorporated into the conventional temperature feedback control, so that the adaptive temperature controller can perform feedforward correction before the temperature deviation has fully occurred.

[0091] The control signal u(t) of the adaptive temperature controller is divided into two parts, namely the feedforward compensation component u. ff (t) and feedback component u fb (t):

[0092] u(t)=u ff (t)+u fb (t)

[0093] Where: u ff (t) is used to apply power adjustment u in advance when significant environmental disturbances are detected. fb (t) then performs conventional closed-loop correction on the actual temperature sampling error;

[0094] To highlight creativity while also considering flexibility, the environmental disturbance compensation amount ΔT can be... pred (t) is passed through the nonlinear mapping operator Ψ(·), and then multiplied by the gain coefficient Ω(t) that can be dynamically adjusted with time to obtain the feedforward signal:

[0095] u ff (t)=Ω(t)Ψ(ΔT pred (t))

[0096] In the formula: Ψ(·) is a designable nonlinear function, which can be selected in a smooth saturated form with two parameters to balance linear response to small deviations and saturation limit for large deviations, and is used to compensate for environmental disturbances ΔT. pred When (t) is large, it is amplified exponentially, with the environmental disturbance compensation amount ΔT pred When (t) is small, apply flexible constraints to avoid overresponding;

[0097] Ω(t) is a time-varying gain that can be adjusted in the range of [0, +∞) according to actual operating conditions (such as the current power limit of the heater or the grid voltage limit) to prevent overshoot or safety hazards in extreme cases;

[0098] The conventional PID or incremental PID controller is modified into an adaptive PID controller, so that its gain coefficient Ω(t) can be gradually corrected during operation.

[0099] Where: the temperature setpoint is denoted as T. set (t), the actual temperature is T meas (t), defining instantaneous temperature error:

[0100] e(t) = T set (t)-T meas (t)

[0101] Example of a feedback term using incremental adaptive PID:

[0102]

[0103] Where: u fb (t k α is the feedback control signal at discrete time t; p (t k ), α i (t k ), α d (t k ) represent the proportional, integral, and differential gains that are updated online over time; Δt = t k+1 -t k The discrete sampling period;

[0104] To enable the controller to sense the impact of environmental disturbances, the environmental disturbance compensation amount ΔT generated in the previous step can also be used. pred (t) is incorporated into the gain update rule, for example, if the environmental disturbance compensation amount ΔT pred If (t) is too large, the differential or proportional gain can be temporarily increased to speed up the response;

[0105] When in use, a feedforward compensation component u is introduced simultaneously. ff (t) and adaptive PID feedback component u fb (t) achieves a double-insurance response to external environmental disturbances: it can intervene in time before or immediately after a sudden change in the environment, and continuously make fine adjustments in the subsequent actual temperature feedback; the adaptive gain mechanism can dynamically adjust the control intensity under different operating conditions, so that the system can avoid serious overshoot or lag even in extreme environments.

[0106] Step 302: Online implementation and closed-loop iteration of the dynamic control strategy

[0107] In embedded controllers or host computer systems, the following calculations are performed according to a fixed period Δ(t):

[0108] Obtain the environmental disturbance compensation amount ΔT at the current time t. pred (tk), actual temperature T meas (t k Temperature setpoint T set (t k and grid voltage V env (t k Information such as feedforward and feedback algorithms are used to calculate the feedforward compensation component u. ff (t k ) and feedback component u fb (t k ), thus obtaining the total output u(t) k The total output u(t) will be... k This is converted into an actual heater power or duty cycle adjustment signal to control the output heat of the high-voltage PTC electric heater.

[0109] The new power signal caused a change in the heater temperature, and the temperature sensor detected the new actual temperature T again. meas (t k+1 The data is then uploaded to step 102 (the sensor network and high-resolution data acquisition unit in steps one and two) and the adaptive temperature controller in this step.

[0110] Meanwhile, the environmental disturbance compensation amount ΔT pred (t) will also be updated in the next sampling period based on the latest environmental data or model predictions (from the online algorithm in step 2); the adaptive PID controller then combines the current error e(t) with the data. k+1 ) and environmental disturbance compensation amount ΔT pred (t k+1 Complete adaptive gain adjustment;

[0111] By forming a closed loop through multiple iterations, when a new large disturbance occurs in the external environment, both feedforward compensation and adaptive gain will quickly adapt to maintain the stability and accuracy of temperature control. To highlight the synergistic effect of the algorithm here, the feedforward and feedback signals can be described in the same expression (taking discrete time as an example):

[0112] u(t k+1 )=Ω(t k )Ψ(ΔT pred (t k ))+u fb (t k+1 )

[0113] Where the feedback component u fb (t k+1 The expression, given by the aforementioned incremental adaptive PID, highlights the feedforward compensation (based on the environmental disturbance compensation amount ΔT). pred (t k Dual regulation of )) and adaptive feedback (based on actual temperature error);

[0114] The final output is an online adaptive dynamic control strategy and a real-time generated control signal u(t), which drives the high-voltage PTC electric heater in practice to achieve control over the set temperature T. set Tracking of (t);

[0115] In use, it forms a closed-loop adaptive control closed loop for actual industrial application scenarios. It has excellent response speed and control accuracy, especially for conditions with high complexity of multi-dimensional environmental disturbances (such as multiple low temperature zones, sudden voltage drops or humidity changes). Through bidirectional linkage with the second step (compensation model), it forms a dynamic strategy of rapid prediction and real-time correction. It uses two levels of information (predicted quantity and current measurement error) to ensure optimal temperature stability.

[0116] Step 4: After the adaptive control strategy outputs the control signal u(t) stably, extreme scenarios are configured in the multi-dimensional environment simulation chamber and feedback data is transmitted back. After completing the multi-scenario acquisition, the model-controller coupling performance is evaluated based on the cumulative temperature deviation functional γ. Long-term verification is performed to continuously optimize the gain parameters.

[0117] Step four includes the following:

[0118] Step 401: Multi-environment scenario configuration and data backhaul

[0119] Using the multi-dimensional environment simulation chamber constructed in the first step, and combining the functions of a programmable temperature source, humidity source, and variable voltage source, several sets of extreme test scenarios are systematically configured. Example scenarios include:

[0120] Extremely low temperature and high humidity start-up scenarios: for example, below -20°C and relative humidity >90%; strong power grid fluctuation scenarios: voltage V at the power supply end. env (t) randomly superimposed high-frequency disturbances or simulated sudden voltage drops; composite scenario: rapid changes in temperature, humidity and voltage together, used to test the robustness of the system under multiple coupled disturbances;

[0121] Using the sensor network and data acquisition unit in step one, the actual temperature T is... meas (t), Temperature setpoint T set (t) Environmental parameters, such as cabin temperature T env (t), Relative humidity inside the cabin H env (t), power supply voltage V env (t), Load current I load The temperature response curves (t) and control signal u(t) are recorded synchronously. This process ensures that all important temperature control-related variables have clear timestamps at different times. Simultaneously, real-time or periodically summarized test data is sent back to the data processing center for calibration analysis in the next step. The final core output data includes temperature response curves T under several scenarios. meas (t), temperature control signal u(t), and various environmental parameter variables (T) env (t),H env (t),V env (t),I load (t));

[0122] When in use, it can capture all elements of the system's data in real or simulated extreme environments, providing sufficient samples and measurement basis for subsequent calibration and verification. Multi-scenario testing can more comprehensively reveal the potential defects or deficiencies of the system under boundary conditions.

[0123] Step 402: Global Error Correction and Model-1 Controller Co-tuning

[0124] To characterize the system during the test time interval [0, T] test The deviation between the setpoint and the actual measured value of temperature is considered, reflecting the combined effect of environmental parameters (temperature, humidity, voltage, etc.) and their dynamic changes on control quality. The following quantities are defined:

[0125] Environment state vector Where t∈[0,T] test ];

[0126] T env (t) represents the external ambient temperature, H env (t) represents the external ambient humidity, V env (t) represents the power supply voltage to the heater;

[0127] Temperature deviation function: δ(t)=T set (t)-T meas (t), where T set (t) represents the set temperature, T meas (t) represents the current measured temperature; time gradient information is introduced to capture the impact of the rate of change of the environment over a short period of time (such as power grid transients, voltage disturbances, etc.) on the control effect, using the norm of the environmental gradient vector:

[0128] Weighting matrices and amplification factors A and B: These are matrices (which can be diagonal or sparse symmetric matrices) that weight the absolute environmental quantity and the rate of change of the environment, respectively, and have the same dimensions as the environmental vector. For example:

[0129] A=diag(a1,a2,a3),B=diag(b1,b2,b3),

[0130] Among them, a i ,b i ≥0, can be configured according to actual testing needs and importance.

[0131] κ1, κ2, and κ3 are all positive nonlinear amplification factors (κ... i >0), used to amplify or reduce the impact of temperature deviation and environmental disturbances on the overall system;

[0132] p,q≥1 represents the norm exponent, which can be a 1-norm, a 2-norm, or other higher-order norms to flexibly adapt to different engineering needs.

[0133] Based on the above definition, the cumulative temperature deviation functional γ is constructed as follows:

[0134]

[0135] Based on the environmental compensation model M established in the second step and its compensation amount ΔT for environmental disturbances. pred If the prediction effect of (t) shows a significant deviation in multi-scenario testing, that is, the cumulative temperature deviation functional γ is higher than expected, the backhaul data can be used for secondary regression or online learning to update, so as to correct the weight vector w or nonlinear mapping function φ(·) of the environmental compensation model M.

[0136] For the adaptive PID controller or incremental controller defined in step three, the initial gain or gain update rule can be reset or adjusted based on indicators such as overshoot and settling time found in the test results. For example, the cumulative temperature deviation functional γ can be compared in different scenarios. If the cumulative temperature deviation functional γ is found to be significantly higher than the average level in a certain scenario, it indicates that the controller's adaptability to that scenario is insufficient, and the gain can be fine-tuned accordingly.

[0137] If the modification of the environmental compensation model results in an environmental disturbance compensation amount ΔT pred If (t) changes significantly, the gain coefficient Ω(t) or the parameter of the nonlinear function Ψ(·) of the feedforward compensation needs to be updated synchronously to ensure that the feedforward and feedback continue to work in synergy and do not cancel each other out or over-superimpose. The final output includes: the updated environmental compensation model M. ′ and its core parameters (such as w) ′ φ ′ (·) etc.), the adjusted adaptive temperature control strategy (including the updated α) p (t), α i (t), α d (parameters such as Ω(t), Ψ(·)).

[0138] In use, by focusing on the analysis of indicators such as cumulative temperature deviation functional Y, an intuitive assessment of the overall control performance of the system can be achieved. It can quickly locate the link with the maximum error or insufficient response in multiple scenarios and dynamically adjust the environmental compensation model and controller gain simultaneously to ensure that high temperature control accuracy can still be maintained under new extreme conditions, avoiding mismatch caused by modifying the model or controller alone.

[0139] Step 403: Long-term running verification and convergence evaluation of the self-learning mechanism

[0140] For the environmental compensation model-PID controller combination that has been adjusted in the previous step, it is necessary to conduct verification of a longer period and a more complex environmental sequence. The simulation chamber can be switched sequentially over a day or longer to test the stability of the system over a longer period of time.

[0141] For example, the cabin temperature is randomly switched between -10℃ and 50℃, accompanied by random voltage disturbances. The system is recorded to see if it can maintain the cumulative temperature deviation functional γ and other indicators at a low level during this process.

[0142] Define a convergence metric Θ(τ) to evaluate the gradual convergence of a self-learning algorithm, such as:

[0143] Θ(τ)=||w(τ)-w(τ-Δτ)|| p +|α p (τ)-α p (τ-Δτ)|+…

[0144] Where: w(τ) is the weight vector of the environmental compensation model at time τ, α p (τ) is the proportional gain of the controller, ||·|| p Represents the p-norm;

[0145] When the convergence index Θ(τ) approaches a minimum value or a stable range, it indicates that the parameter updates of the compensation model and the controller have nearly converged, and the system is in a steady state in the dynamic environment.

[0146] If the convergence index Θ(τ) continues to oscillate or rise during long-term testing, it indicates that the system cannot learn appropriate compensation or gain under certain operating conditions, and further modifications to the algorithm or constraints are needed, i.e., self-learning convergence index, such as reducing the learning step size or limiting the parameter range.

[0147] The temperature response curve T collected during the long-term test meas The cumulative temperature deviation functional Υ and convergence index Θ(τ) are visualized and comprehensively analyzed in the data processing center to determine whether the system has achieved the expected industrial application indicators (such as temperature stability, power consumption and safety protection).

[0148] If long-term verification shows that the system indicators meet or exceed expectations, the overall solution enters the practical application stage; if there are still shortcomings, the model and controller strategies in the second and third steps can be further iterated and optimized based on the verification results.

[0149] During use, long-term, multi-round, and cross-scenario testing was conducted to ensure that the system maintains good temperature control performance under different day-night temperature gradients and power grid load changes, and to verify the continuous effectiveness of the adaptive strategy. The convergence evaluation provides a key reference for judging whether this solution can operate reliably in actual engineering in the long term, and also accumulates real data and experience for possible large-scale applications in the future.

[0150] Through continuous testing and iterative calibration under simulated extreme environments, this solution demonstrates significantly superior temperature control accuracy and safety compared to traditional technologies in practical industrial applications, achieving adaptive compensation and long-term reliable operation against unconventional environmental disturbances. Close coordination and full information sharing among all components (environmental simulation chamber, multi-channel data processing, adaptive control algorithm, and dynamic calibration mechanism) ensure high reliability of the entire solution in dealing with harsh environments and multidimensional dynamic disturbances.

[0151] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0152] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0153] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0154] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0155] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An electronic functional testing system for a high-voltage PTC electric heater, characterized in that: include, When it is detected that the high-voltage PTC electric heater needs to simulate an extreme environment, the multi-dimensional environment simulation chamber performs programmable temperature and humidity control and variable voltage source operation, periodically adjusts and collects environmental parameters in real time, and generates multi-source basic data containing extreme disturbances; the environmental parameters include chamber temperature, chamber relative humidity, power supply voltage and load current; After data aggregation, wavelet filtering and feature extraction algorithms are applied to the environmental parameter data to generate denoised and normalized signals and construct an environmental compensation model to output the environmental disturbance compensation amount. First, all original signals are time-base aligned, and an adaptive filtering strategy based on wavelet transform is used to decompose each channel signal at different scales. After filtering in different frequency bands, the signals are reconstructed, key features are extracted and recorded as multi-channel feature vectors, and multi-channel signals are obtained. A high-dimensional input vector is formed by combining the multi-channel signals, the components of the multi-channel feature vectors, and the comprehensive environmental value. An environmental compensation model is constructed based on this high-dimensional input vector, outputting the environmental disturbance compensation amount for temperature deviations. The comprehensive environmental value is used to uniformly quantify the combined environmental state and dynamic changes formed by temperature, humidity, and power supply voltage, quickly reflecting the overall severity of the current test environment and prompting the system to increase temperature control alerts and compensation efforts when the value rises rapidly. When the environmental compensation model outputs the environmental disturbance compensation amount in real time and detects that the temperature deviation is greater than expected, the adaptive temperature control algorithm combines the feedforward and feedback paths, and superimposes the corresponding disturbance information in the online adjustment of PID gain to form a corrected power control signal; the control signal of the adaptive temperature controller is divided into feedforward compensation component and feedback component, and the environmental disturbance compensation amount is multiplied by the gain coefficient after passing through the nonlinear mapping operator to obtain the feedforward signal. When new environmental disturbance data appears, multi-scenario tests are performed in the environmental simulation chamber and the temperature response curve and environmental disturbance compensation amount are recorded. Based on the cumulative temperature deviation evaluation results, the model weight and control gain parameters are optimized simultaneously. The feedforward compensation component and feedback component are calculated according to the feedforward algorithm and the feedback algorithm to obtain the total output. The total output is converted into the actual heater power or duty cycle adjustment signal to control the output heat of the high-voltage PTC electric heater.

2. The electronic function testing system for the high-voltage PTC electric heater according to claim 1, characterized in that: The simulation chamber is equipped with programmable heat and cold sources, an adjustable humidity generator, and a variable voltage source. By selecting composite materials resistant to temperature difference shock, heat insulation and isolation layers are laid out inside, and an air duct system is added. Inside the simulation chamber, a sensor array is constructed according to high temperature zone, low temperature zone, and strong convection zone to collect environmental parameter components and construct an overall environmental value. When the overall environmental value rises rapidly, an alert is sent to the outside.

3. The electronic function testing system for the high-voltage PTC electric heater according to claim 2, characterized in that: An online learning method is used to iteratively update the weight vector of the environmental compensation model.

4. The electronic function testing system for the high-voltage PTC electric heater according to claim 3, characterized in that: The conventional PID or incremental PID controller is transformed into an adaptive PID controller, so that its gain coefficient can be gradually corrected during operation, and the environmental disturbance compensation is incorporated into the gain update rule.

5. The electronic function testing system for the high-voltage PTC electric heater according to claim 4, characterized in that: The environmental disturbance compensation amount is updated based on the latest environmental data or model predictions. The adaptive PID controller then combines the current error and the environmental disturbance compensation amount to complete the adaptive adjustment of the gain.

6. The electronic function testing system for the high-voltage PTC electric heater according to claim 5, characterized in that: After configuring several sets of extreme test scenarios, the environmental parameters and control signals are recorded synchronously, and the temperature response curves, temperature control signals and various environmental parameter variables under several scenarios are output. A cumulative temperature deviation functional is constructed based on the environmental compensation model and its predictive effect on environmental disturbance compensation. If the cumulative temperature deviation functional is higher than expected, the backhaul data is used for quadratic regression or online learning updates to correct the weight vector or nonlinear mapping function of the environmental compensation model.

7. The electronic function testing system for the high-voltage PTC electric heater according to claim 6, characterized in that: For adaptive PID controllers, the initial gain is reset or the gain update rule is adjusted based on the overshoot and settling time found in the test results. If the modification of the environmental compensation model causes a significant change in the amount of environmental disturbance compensation, the gain coefficient or parameters of the nonlinear function of the feedforward compensation need to be updated simultaneously to obtain the adjusted adaptive temperature control strategy.

8. The electronic function testing system for the high-voltage PTC electric heater according to claim 7, characterized in that: For the adjusted environmental compensation model-PID controller combination, a convergence index is defined to evaluate the gradual convergence of the self-learning algorithm. If the convergence index continues to oscillate or rises during long-term testing, the self-learning convergence index is modified.

9. The electronic function testing system for the high-voltage PTC electric heater according to claim 8, characterized in that: The temperature response curves, cumulative temperature deviation functionals, and convergence indices collected during long-term testing are visualized and comprehensively analyzed to determine whether the expected industrial application indicators have been achieved. If long-term verification shows that the industrial application indicators do not meet expectations, further iterative optimization can be performed based on the verification results.

Images