Intelligent control system and method for switchgear

Abstract

The application provides an intelligent control system and method of a switching device, and relates to the technical field of electrical switching, comprising a multi-source sensing module: used for collecting contact current signals, contact voltage signals and contact displacement signals in real time during opening and closing operations of the switching device, and performing filtering and preprocessing on the collected signals; connected with the multi-source sensing module, used for calculating arc energy evaluation values in the opening process in real time based on a pre-established nonlinear mapping relationship dynamically corrected by environment temperature and humidity, according to the preprocessed contact current signals and contact displacement signals; wherein the nonlinear mapping relationship takes the current change rate at the current zero-crossing moment, the contact opening distance at the moment and the main loop peak current as input variables, performs rolling optimization by using a model predictive control algorithm, can dynamically generate an optimal opening speed curve under the premise of meeting various physical constraints, and thus actively suppresses and minimizes the arc energy.

Description

Technical Field

[0001] This invention relates to the field of electrical switch technology, specifically to an intelligent control system and method for switchgear. Background Technology

[0002] Switchgear such as high-voltage circuit breakers and medium-voltage vacuum circuit breakers are responsible for carrying and breaking normal load current and quickly clearing fault current in power systems. Among them, the opening operation is an extreme process of energy conversion and dissipation. If the electric arc generated in the contact gap when the operating mechanism drives the contacts to separate cannot be effectively controlled, it will lead to severe contact erosion, a decrease in the insulation performance of the arc extinguishing medium, and in extreme cases, even break-out failure, causing equipment damage or system accidents.

[0003] Currently, the existing technical solutions for controlling the tripping operation of switchgear mainly adopt a fixed-sequence open-loop control strategy. That is, according to the factory settings of the switchgear, the operating mechanism is driven to complete the tripping action according to a predetermined single speed-stroke characteristic curve. This method is simple to implement, but its control parameters (such as coil drive current and tripping speed curve) are set under specific standard test environments (such as 20℃ and 50% relative humidity) and rated operating conditions, which cannot adapt to the changes in dynamic factors such as load current, ambient temperature and humidity, and performance degradation of the mechanism in actual operation.

[0004] To address the shortcomings in existing switchgear control technologies, such as insufficient decoupling control of arc energy and opening speed curves, and the lack of dynamic correction mechanisms for arc characteristic drift caused by changes in ambient temperature and humidity, this invention provides an intelligent control system for switchgear. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent control system for switchgear. This control system is used to control the operating mechanism to perform opening operations, and includes a multi-source sensing module: used to collect contact current signals, contact voltage signals and contact displacement signals in real time during the opening and closing operations of the switchgear, and to perform filtering and preprocessing on the collected signals.

[0006] Arc energy assessment module: connected to the multi-source sensing module, used to calculate the arc energy assessment value during the opening process in real time based on the pre-processed contact current signal and contact displacement signal, and on a pre-established nonlinear mapping relationship dynamically corrected by ambient temperature and humidity; wherein the nonlinear mapping relationship uses the current change rate at the current zero crossing moment, the contact opening distance at that moment, and the peak current of the main circuit as input variables.

[0007] An adaptive control decision module, connected to the arc energy assessment module, is used to take the arc energy assessment value as feedback quantity and minimize the arc energy as the control objective. Within a preset speed-stroke constraint space, it uses a model predictive control algorithm to perform rolling optimization iterative solution to generate the optimal opening speed reference curve and the corresponding initial coil drive control parameters, and outputs the initial coil drive control parameters to the operating mechanism.

[0008] Calibration-free degradation compensation module: connected to the adaptive control decision module, used to calculate the performance degradation trend index of the operating mechanism based on the action characteristic parameter sequence of the historical opening and closing actions of the switchgear, using an exponential weighted moving average algorithm, and generate the control parameter compensation amount based on the pre-stored mapping relationship between the degradation degree and the control parameter compensation amount;

[0009] Intelligent control execution module: used to superimpose the control parameter compensation amount onto the optimal opening speed reference curve and the corresponding coil drive control parameters; generate the final drive command to control the operating mechanism to perform the opening operation, and collect the coil current waveform and the actual contact stroke curve in real time, and feed the collected data back to the multi-source sensing module and the calibration-free degradation compensation module.

[0010] Preferably, the nonlinear mapping relationship is established in the following manner:

[0011] Offline calibration stage: Based on the rated voltage level and arc extinguishing medium type of the switchgear, the tripping waveform data under different operating conditions are obtained through high voltage breaking tests, and the basic mapping coefficients are obtained by least squares polynomial regression fitting.

[0012] During the online operation phase: Based on real-time collected ambient temperature and relative humidity, the basic mapping coefficients are linearly and dynamically corrected using an environmental compensation factor to obtain effective coefficients for calculating the arc energy assessment value. .

[0013] Preferably, the effective coefficient for calculating the arc energy assessment value is obtained using an environmental compensation factor, wherein the ambient temperature and relative humidity are collected in real time by a PT100 and a capacitive humidity sensor:

[0014] ,

[0015] The effective coefficient is after environmental dynamic correction; The basic mapping parameters for offline calibration are: T is the real-time ambient temperature; H is the real-time relative humidity. Reference temperature; Reference humidity; This refers to the temperature sensitivity coefficient. This is the humidity sensitivity coefficient.

[0016] Preferably, the specific calculation method for the arc energy assessment value during the circuit breaker tripping process is as follows:

[0017] In the offline phase based on nonlinear mapping relationships, tripping waveform data of different voltage levels and arc-extinguishing dielectric cartridges are acquired through a high-voltage breaking test bench. In the online phase, the moment of current zero-crossing is identified. Based on the discrete sampled data in the neighborhood at that moment, the absolute value of the rate of change of current is calculated using a numerical differentiation algorithm. Based on the real-time travel data measured by the displacement sensor, combined with the current zero-crossing time The timestamp is used to obtain the contact opening distance at that moment through an interpolation algorithm. The maximum instantaneous absolute value of the main circuit current during this tripping operation is obtained as the peak current of the main circuit. The arc energy assessment value is obtained by combining the absolute value of the current change rate, the contact gap, and the peak current of the main circuit. This arc energy assessment value is then labeled as... .

[0018] Preferably, the adaptive control decision module solves for the optimal tripping speed reference curve as follows:

[0019] First, a comprehensive cost function is constructed, which includes an arc energy penalty term, a velocity tracking error term, and a control variable change rate penalty term. This comprehensive cost function is then labeled as... ;

[0020] Subsequently, the model predictive control algorithm, within each fixed microsecond-level control cycle, based on the aforementioned comprehensive cost function... A constrained online rolling optimization solution is performed, and the solution process is subject to the following constraints: peak value constraint of coil drive current, maximum allowable acceleration constraint of contacts, and upper and lower limits constraint of duty cycle of power drive signal.

[0021] Preferably, the model predictive control algorithm performs the following steps within each fixed microsecond-level control cycle:

[0022] S51: Based on the predictive model of the operating mechanism, predict the displacement trajectory, velocity trajectory and corresponding arc energy assessment value of the contact under different candidate driving parameters in the future control cycle.

[0023] S52: Under the condition that the above constraints are satisfied at the same time, the optimal control sequence is obtained by minimizing the preset comprehensive cost function through an online quadratic programming solver;

[0024] S53: Output the first element of the optimal control sequence as the coil drive control parameter for the current control cycle;

[0025] S54: When the next control cycle arrives, the prediction model is corrected by using the real-time feedback data collected by the multi-source sensing module, and steps S51 to S53 are repeated to achieve rolling optimization closed-loop control.

[0026] The prediction model is a discrete-time state-space model of the operating mechanism, whose state variables include at least coil current, contact movement speed and contact displacement, and the control variable is the duty cycle of the PWM drive signal. The model is obtained by derivation based on physical mechanisms or by system identification methods based on input and output data, and the model parameters are corrected online using real-time feedback data during operation.

[0027] Preferably, the adaptive control decision module performs a validity assessment of the solution results of the model predictive control algorithm in each control cycle, and the assessment criteria include:

[0028] (1) When the online quadratic programming solver returns an infeasible state, it is determined that the solution result exceeds the preset constraint boundary;

[0029] (2) If the optimal control quantity calculated within three consecutive control cycles touches the upper or lower limit of the constraint boundary, and the arc energy assessment value does not show a downward trend, it is determined that the solution result has not converged within a consecutive preset number of cycles.

[0030] When any of the above judgment conditions are met, the adaptive control decision module triggers a preset safety protection control strategy.

[0031] Preferably, the safety baseline control strategy includes:

[0032] First control action: Stop using the optimal tripping speed reference curve generated in real time by the model predictive control algorithm, and instead load the reference tripping speed curve pre-stored in the non-volatile memory; the reference tripping speed curve is obtained by fitting the opening and closing data of this type of switchgear under various specified operating conditions in type tests;

[0033] The second control action is to forcibly switch the control mode from closed-loop feedback adjustment mode to open-loop fixed duty cycle drive mode, and output drive signals according to the preset time-duty cycle table corresponding to the reference opening speed curve until the opening position signal is detected.

[0034] Preferably, the calibration-free degradation compensation module specifically includes:

[0035] Feature extraction unit: used to extract action feature parameters, including action time, average stroke speed, stroke end overshoot, and coil current peak value, from the coil current waveform and the actual contact stroke curve of each opening and closing operation;

[0036] Degradation trend calculation unit: It is used to process the numerical sequence of action feature parameters of the same type in multiple consecutive operations using the exponential weighted moving average algorithm to calculate the degradation trend index after the current operation. The deviation of the feature parameters of the recent operation from the reference benchmark value is given the first weight, and the historical cumulative degradation trend is given the second weight.

[0037] Compensation amount generation unit: used to input the calculated degradation trend index into the pre-generated mapping relationship between degradation degree and control parameter compensation amount, to obtain the control parameter compensation amount required for the current operating cycle;

[0038] The mapping relationship between the degree of degradation and the compensation amount of the control parameters is established in advance in the following way:

[0039] Accelerated aging tests were conducted on prototype switchgear of the same model. During the test, the changes in its action characteristic parameters relative to the initial state were measured periodically, and the increment of control parameters required to restore the action characteristic parameters to near the initial state was determined simultaneously. The mapping relationship was generated by fitting multiple sets of changes with corresponding increment data.

[0040] An intelligent control method for switchgear, the method employing the aforementioned intelligent control system for switchgear, includes the following steps:

[0041] S1: Acquire contact current signal, contact voltage signal and contact displacement signal, and perform filtering preprocessing;

[0042] S2: Identify the moment when the current crosses zero, extract the current change rate, the contact opening distance, and the peak current of the main circuit at that moment, and combine them with the nonlinear mapping relationship after dynamic correction by ambient temperature and humidity to calculate the arc energy assessment value during the opening process in real time.

[0043] S3: Using the calculated arc energy assessment value as the feedback quantity and minimizing the arc energy as the control objective, within the preset speed-stroke constraint space, a model predictive control algorithm is used to perform rolling optimization iterative solution to generate the optimal opening speed reference curve and the corresponding initial coil drive control parameters.

[0044] S4: Based on the sequence of action characteristic parameters of each opening and closing operation, calculate the performance degradation trend index of the operating mechanism using the exponential weighted moving average algorithm, and generate the control parameter compensation amount based on the pre-stored mapping relationship between the degree of degradation and the control parameter compensation amount.

[0045] S5: The control parameter compensation amount is superimposed on the initial coil drive control parameters to generate the final drive command to control the operating mechanism to perform the opening operation; the coil current waveform and actual contact stroke curve of this opening operation are collected to update the degradation trend index and serve as the basis for the next control calculation.

[0046] This invention provides an intelligent control system and method for switchgear. It has the following beneficial effects:

[0047] By acquiring key signals in real time through multi-source sensing modules and based on an arc energy assessment model dynamically corrected for environmental conditions, the system achieves accurate and real-time assessment of arc energy during the opening process. Furthermore, using this assessment value as feedback, the system employs a model predictive control algorithm for rolling optimization. This enables the dynamic generation of the optimal opening speed curve while satisfying various physical constraints, thereby achieving proactive suppression and minimization of arc energy. This significantly reduces contact erosion, extends the electrical life of the switchgear, and improves the stability and consistency of its opening and closing process.

[0048] An online environmental temperature and humidity compensation mechanism and a calibration-free mechanism degradation compensation strategy were introduced. By real-time correction of the evaluation model parameters, the drift effect caused by environmental changes on arc characteristics was effectively overcome, ensuring the applicability of the control strategy under different operating conditions. At the same time, the system automatically identifies the performance degradation trend of the operating mechanism by analyzing historical operating data and generates corresponding control parameter compensation amounts, realizing adaptive compensation for factors such as mechanism aging and wear. This enables the switchgear to maintain excellent and stable tripping performance throughout its entire life cycle, reducing maintenance needs and reliance on periodic calibration.

[0049] By using an adaptive control decision module to monitor and assess the effectiveness of the optimization solution in real time, a safety-guaranteed control strategy can be immediately triggered if an anomaly is detected or the control effect fails to meet expectations. This strategy can seamlessly switch to an open-loop drive mode based on a pre-stored baseline curve, ensuring that the tripping operation can be reliably and completely executed under any complex or abnormal conditions. This fundamentally avoids the risk of mechanical jamming or failure to operate due to control algorithm failure, greatly improving the operational safety and reliability of the entire system. Attached Figure Description

[0050] Figure 1 This is a system flowchart of the present invention;

[0051] Figure 2 This is a schematic diagram of feature extraction and exponential weighted trend calculation for the calibration-free degradation compensation module of the present invention;

[0052] Figure 3 This is a flowchart of the method of the present invention;

[0053] Figure 4 This is a logic diagram for the switching of the solution anomaly detection and safety backup strategy in this invention. Detailed Implementation

[0054] 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.

[0055] like Figures 1 to 4 As shown, the present invention proposes an intelligent control system for switchgear. This control system is used to control operating mechanisms such as electromagnets and spring mechanisms to perform opening operations. It includes a multi-source sensing module for real-time acquisition of contact current signals, contact voltage signals and contact displacement signals during the opening and closing operations of the switchgear, and for filtering and preprocessing the acquired signals.

[0056] Arc energy assessment module: connected to the multi-source sensing module, used to calculate the arc energy assessment value during the opening process in real time based on the pre-processed contact current signal and contact displacement signal, and on a pre-established nonlinear mapping relationship dynamically corrected by ambient temperature and humidity; the nonlinear mapping relationship uses the current change rate at the current zero crossing moment, the contact opening distance at that moment, and the peak current of the main circuit as input variables.

[0057] The nonlinear mapping relationship is established in the following way:

[0058] Offline calibration stage: Based on the rated voltage level and arc extinguishing medium type of the switchgear, the tripping waveform data under different operating conditions are obtained through high voltage breaking tests, and the basic mapping coefficients are obtained by least squares polynomial regression fitting.

[0059] During the online operation phase: Based on real-time collected ambient temperature and relative humidity, the basic mapping coefficients are linearly and dynamically corrected using an environmental compensation factor to obtain effective coefficients for calculating the arc energy assessment value. ;

[0060] Ambient temperature and relative humidity are collected in real time using a PT100 and a capacitive humidity sensor.

[0061] ;

[0062] The effective coefficient is after environmental dynamic correction; The basic mapping parameters for offline calibration are: T is the real-time ambient temperature; H is the real-time relative humidity. Reference temperature; Reference humidity; This refers to the temperature sensitivity coefficient. Humidity sensitivity coefficient;

[0063] Specifically, among them The value is set to 20℃. The value is 50%. Temperature sensitivity coefficient Humidity sensitivity coefficient (e.g., in a vacuum interrupter) , In the SF6 arc extinguishing chamber , This compensation formula can avoid evaluation errors caused by the drift of arc characteristics due to changes in temperature and humidity.

[0064] The specific calculation method for the arc energy assessment value during the circuit breaker tripping process is as follows:

[0065] In the offline phase based on nonlinear mapping relationships, tripping waveform data of different voltage levels and arc-extinguishing dielectric cartridges are acquired through a high-voltage breaking test bench. In the online phase, the moment of current zero-crossing is identified. Based on the discrete sampled data in the neighborhood at that moment, the absolute value of the rate of change of current is calculated using a numerical differentiation algorithm. Based on the real-time travel data measured by the displacement sensor, combined with the current zero-crossing time The timestamp is used to obtain the contact opening distance at that moment through an interpolation algorithm. The maximum instantaneous absolute value of the main circuit current during this tripping operation is obtained as the peak current of the main circuit. The arc energy assessment value is obtained by integrating the absolute value of the current change rate, the contact gap, and the peak current of the main circuit. This arc energy assessment value is then labeled as... .

[0066] Specifically, during the offline calibration phase, the nonlinear mapping evaluation model obtains the opening waveform data of the arc-extinguishing medium (vacuum, SF6, environmentally friendly gas) box at different voltage levels (such as 12kV, 24kV, 40.5kV) through a high-voltage breaking test bench.

[0067] Furthermore, the current acquisition unit in the multi-source sensing module performs analog-to-digital conversion on the primary loop current at a sampling rate of no less than 100kHz, generating a discrete current sequence. A zero-crossing comparison algorithm is then used to identify the precise moments of current alternation between positive and negative values. The slope of the current sampling points in the neighborhood of the zero-crossing point is estimated by using a least squares linear fitting algorithm based on a sliding window, thereby calculating the absolute value of the rate of change of the current at the moment of zero crossing.

[0068] Simultaneously, the displacement acquisition unit monitors the movement position of the main shaft of the operating mechanism in real time using a high-precision rotary encoder or linear grating ruler. The arc energy assessment module utilizes a linear interpolation algorithm based on precise timing... The timestamp is used to interpolate between two adjacent sampling points recorded by the displacement sensor, reconstructing and obtaining the contact opening distance corresponding to that moment. Finally, the current conditioning circuit in the multi-source sensing module includes a hardware peak hold circuit. This peak hold circuit is reset and activated when the controller issues a tripping command to capture the maximum instantaneous absolute value of the tripping current waveform from the transient state to before it crosses zero. The processor reads this holding voltage through an analog-to-digital converter and calculates it to obtain the peak current of the main circuit for this operation. As an alternative, a digital peak detection algorithm can be used to compare the current sampling sequence during the circuit breaker tripping process point by point to find the maximum absolute value. ;

[0069] but ;

[0070] The effective coefficient is after environmental dynamic correction;

[0071] "For example, in a 12kV vacuum circuit breaker breaking test, the absolute value of the current change rate at the moment of current zero crossing was measured to be 15A / µs, the contact opening distance was 3mm, the peak current of the main circuit was 20kA, and the basic coefficient was..." The effective coefficient is 0.02 after temperature and humidity correction. The value is 0.021. Substituting this value into the formula, we obtain the current estimated arc energy. (Unit: Joule equivalent value); This value accurately reflects the intensity of the arcing process.

[0072] The adaptive control decision module, connected to the arc energy assessment module, uses the arc energy assessment value as feedback quantity and minimizes arc energy as the control objective. Within the preset speed-stroke constraint space, it employs a model predictive control algorithm to perform rolling optimization iterative solution, generating the optimal opening speed reference curve and the corresponding initial coil drive control parameters, and outputs the initial coil drive control parameters to the operating mechanism.

[0073] The adaptive control decision module includes the following steps when solving for the optimal tripping speed reference curve:

[0074] First, a comprehensive cost function is constructed, which includes an arc energy penalty term, a velocity tracking error term, and a control variable change rate penalty term. ;

[0075] Specifically, the comprehensive cost function The formula for construction is:

[0076] ;

[0077] in For the predicted time domain length, it represents the number of steps to predict into the future, for example, taking... If the control period is 50µs, then the prediction window is 8 x 50 = 0.4ms; This is the standard MPC notation;

[0078] In order to predict the square of the arc energy assessment value that the switching device will generate in the i-th step in the current k-th control cycle, using the discrete-time state-space model of the operating mechanism;

[0079] The actual speed of the contact is derived from the dynamic model of the operating mechanism. The reference velocity generated for the current iteration; This refers to the change in the control quantity (PWM duty cycle or drive voltage); , , The weighting coefficient has a range of values. ; , ;

[0080] The optimizer employs an online quadratic programming solver, such as OSQP, which returns the optimal control sequence within a 50-second period. If OSQP returns infeasible or D(k) reaches its maximum or minimum value for three consecutive periods, and... If convergence fails, a safety protection strategy is triggered, immediately loading the pre-stored benchmark tripping speed curve and switching to open-loop fixed duty cycle drive to ensure smooth tripping action and avoid mechanism jamming or contact bounce.

[0081] Subsequently, the model predictive control algorithm, within each fixed microsecond-level control cycle, is based on the comprehensive cost function. The online rolling optimization with constraints is performed. The solution process is subject to the following constraints: peak value constraint of coil drive current, maximum allowable acceleration constraint of contact, and upper and lower limit constraints of duty cycle of power drive signal.

[0082] The model predictive control algorithm performs the following steps within each fixed microsecond-level control cycle:

[0083] S51: Based on the predictive model of the operating mechanism, predict the displacement trajectory, velocity trajectory and corresponding arc energy assessment value of the contact under different candidate driving parameters in the future control cycle.

[0084] S52: Under the condition that the above constraints are satisfied at the same time, the optimal control sequence is obtained by minimizing the preset comprehensive cost function through an online quadratic programming solver;

[0085] S53: Output the first element of the optimal control sequence as the coil drive control parameter for the current control cycle;

[0086] S54: When the next control cycle arrives, the prediction model is corrected by using the real-time feedback data collected by the multi-source sensing module, and steps S51 to S53 are repeated to achieve rolling optimization closed-loop control.

[0087] The prediction model is a discrete-time state-space model of the operating mechanism. Its state variables include at least coil current, contact speed and contact displacement, and the control variable is the duty cycle of the PWM drive signal. The model is obtained through derivation based on physical mechanisms or system identification methods based on input and output data, and the model parameters are corrected online using real-time feedback data during operation.

[0088] The effect of imposing this constraint is to ensure that the generated driver instructions do not cause hardware damage or abnormal movement.

[0089] The adaptive control decision module evaluates the effectiveness of the model predictive control algorithm's solution results within each control cycle. The evaluation criteria include:

[0090] (1) When the online quadratic programming solver returns an infeasible state, it is determined that the solution result exceeds the preset constraint boundary;

[0091] (2) If the optimal control quantity calculated within three consecutive control cycles touches the upper or lower limit of the constraint boundary and the arc energy assessment value does not show a downward trend, it is determined that the solution result has not converged within the preset number of consecutive cycles.

[0092] When any of the above judgment conditions are met, the adaptive control decision module triggers the preset safety protection control strategy.

[0093] Safety margin control strategies include:

[0094] First control action: Stop using the optimal tripping speed reference curve generated in real time by the model predictive control algorithm, and instead load the reference tripping speed curve pre-stored in non-volatile memory; the reference tripping speed curve is obtained by fitting the opening and closing data of this type of switchgear under various specified operating conditions in type tests;

[0095] Second control action: Force the control mode to switch from closed-loop feedback adjustment mode to open-loop fixed duty cycle drive mode, and output drive signal according to the preset time-duty cycle table corresponding to the reference opening speed curve until the opening position signal is detected.

[0096] Calibration-free degradation compensation module: Connected to the adaptive control decision module, it is used to calculate the performance degradation trend index of the operating mechanism based on the action characteristic parameter sequence of the historical opening and closing actions of the switchgear, using an exponential weighted moving average algorithm, and generate the control parameter compensation amount based on the pre-stored mapping relationship between the degradation degree and the control parameter compensation amount.

[0097] The calibration-free degradation compensation module specifically includes:

[0098] Feature extraction unit: used to extract action feature parameters, including action time, average stroke speed, stroke end overshoot, and coil current peak value, from the coil current waveform and actual contact stroke curve of each opening and closing operation;

[0099] Degradation trend calculation unit: It is used to process the numerical sequence of action feature parameters of the same type in multiple consecutive operations using the exponential weighted moving average algorithm to calculate the degradation trend index after the current operation. The deviation of the feature parameters of the recent operation from the reference benchmark value is given the first weight, and the historical cumulative degradation trend is given the second weight.

[0100] The mapping relationship between the degree of degradation and the compensation amount of the control parameters is established in advance in the following way:

[0101] Accelerated aging tests were conducted on the same type of switchgear prototype. During the test, the changes in its action characteristic parameters relative to the initial state were measured periodically, and the increment of control parameters required to restore the action characteristic parameters to near the initial state was determined simultaneously. By fitting multiple sets of changes with corresponding increment data, a mapping relationship was generated.

[0102] Compensation quantity generation unit: used to input the calculated degradation trend index into the pre-generated mapping relationship between degradation degree and control parameter compensation quantity, to obtain the control parameter compensation quantity required for the current operating cycle;

[0103] Specifically: Feature extraction unit: Specific parameters include action time, peak coil current, average stroke speed, and end-stroke overshoot. Extraction method: Real-time analysis of waveforms and curves is performed through edge computing terminals, such as peak detection, zero-crossing detection, and numerical integration.

[0104] Next, the degradation trend is calculated, which transforms the single action characteristic parameters into an indicator that can smoothly reflect the long-term trend. The exponentially weighted moving average algorithm is as follows:

[0105] ;

[0106] This is the current degradation trend index calculated after the k-th tripping operation. It is a dimensionless relative value. Positive numbers indicate performance degradation, such as longer time, while negative numbers indicate abnormality, such as abnormally shortened time. This represents the value of a certain feature parameter actually extracted in the k-th operation.

[0107] This is the reference value for the k-th operation;

[0108] This refers to the relative deviation rate of the action parameters compared to the baseline value.

[0109] This is a smoothing factor, with a value range of 0-1. This is a weighted term representing the historical cumulative trend of degradation. This is a historical degradation trend indicator following the previous operation;

[0110] The mapping relationship between the degree of degradation and the amount of compensation was obtained by conducting accelerated aging tests on the same type of switchgear in advance.

[0111] In a laboratory setting, accelerated life tests are conducted on one or more prototype machines of the same model, such as tens of thousands of continuous operations. The degradation of their operational characteristic parameters is measured periodically, and the compensation amount of control parameters required to maintain the original tripping characteristics, such as arcing time and speed curves, is tested simultaneously. A corresponding relationship table is obtained through data fitting. During online operation, the system will use the parameters calculated in step 2... By substituting the values ​​into the corresponding relationship table, the required control parameter compensation amount for the current operating cycle can be obtained.

[0112] Intelligent control execution module: It is used to superimpose the control parameter compensation amount onto the optimal opening speed reference curve and the corresponding coil drive control parameters; generate the final drive command to control the operating mechanism to perform the opening operation, and collect the coil current waveform and the actual contact stroke curve in real time, and feed the collected data back to the multi-source sensing module and the calibration-free degradation compensation module.

[0113] An intelligent control method for switchgear, which employs the aforementioned intelligent control system for switchgear, includes the following steps:

[0114] S1: Acquire contact current signal, contact voltage signal and contact displacement signal, and perform filtering preprocessing;

[0115] S2: Identify the moment when the current crosses zero, extract the current change rate, the contact opening distance, and the peak current of the main circuit at that moment, and combine them with the nonlinear mapping relationship after dynamic correction by ambient temperature and humidity to calculate the arc energy assessment value during the opening process in real time.

[0116] S3: Using the calculated arc energy assessment value as the feedback quantity and minimizing the arc energy as the control objective, within the preset speed-stroke constraint space, the model predictive control algorithm is used to perform rolling optimization iterative solution to generate the optimal opening speed reference curve and the corresponding initial coil drive control parameters.

[0117] S4: Based on the sequence of action characteristic parameters of each opening and closing operation, calculate the performance degradation trend index of the operating mechanism using the exponential weighted moving average algorithm, and generate the control parameter compensation amount based on the pre-stored mapping relationship between the degree of degradation and the control parameter compensation amount.

[0118] S5: The control parameter compensation is superimposed on the initial coil drive control parameters to generate the final drive command to control the operating mechanism to perform the opening operation; the coil current waveform and actual contact stroke curve of this opening operation are collected to update the degradation trend index and serve as the basis for the next control calculation.

[0119] The threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value, it is acceptable.

[0120] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0121] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent control system for switchgear, wherein the control system is used to control the operating mechanism to perform a tripping operation, characterized in that, It includes a multi-source sensing module, an arc energy assessment module, an adaptive control decision module, a calibration-free degradation compensation module, and an intelligent control execution module: Multi-source sensing module: used to collect contact current signal, contact voltage signal and contact displacement signal in real time during the opening and closing operation of the switchgear, and to perform filtering and preprocessing on the collected signals; Arc energy assessment module: connected to the multi-source sensing module, used to calculate the arc energy assessment value during the opening process in real time based on the pre-processed contact current signal and contact displacement signal, and on a pre-established nonlinear mapping relationship dynamically corrected by ambient temperature and humidity; wherein the nonlinear mapping relationship uses the current change rate at the current zero crossing moment, the contact opening distance at that moment, and the peak current of the main circuit as input variables. An adaptive control decision module, connected to the arc energy assessment module, is used to take the arc energy assessment value as feedback quantity and minimize the arc energy as the control objective. Within a preset speed-stroke constraint space, it uses a model predictive control algorithm to perform rolling optimization iterative solution to generate the optimal opening speed reference curve and the corresponding initial coil drive control parameters, and outputs the initial coil drive control parameters to the operating mechanism. Calibration-free degradation compensation module: connected to the adaptive control decision module, used to calculate the performance degradation trend index of the operating mechanism based on the action characteristic parameter sequence of the historical opening and closing actions of the switchgear, using an exponential weighted moving average algorithm, and generate the control parameter compensation amount based on the pre-stored mapping relationship between the degradation degree and the control parameter compensation amount; Intelligent control execution module: used to superimpose the control parameter compensation amount onto the optimal opening speed reference curve and the corresponding coil drive control parameters; generate the final drive command to control the operating mechanism to perform the opening operation, and collect the coil current waveform and the actual contact stroke curve in real time, and feed the collected data back to the multi-source sensing module and the calibration-free degradation compensation module.

2. The intelligent control system for the switchgear according to claim 1, characterized in that: The nonlinear mapping relationship is established in the following way: Offline calibration stage: Based on the rated voltage level and arc extinguishing medium type of the switchgear, the tripping waveform data under different operating conditions are obtained through high voltage breaking tests, and the basic mapping coefficients are obtained by least squares polynomial regression fitting. During the online operation phase: Based on real-time collected ambient temperature and relative humidity, the basic mapping coefficients are linearly and dynamically corrected using an environmental compensation factor to obtain effective coefficients for calculating the arc energy assessment value. .

3. The intelligent control system for the switchgear according to claim 2, characterized in that: The effective coefficient for calculating the arc energy assessment value is obtained using an environmental compensation factor, wherein the ambient temperature and relative humidity are collected in real time by a PT100 and a capacitive humidity sensor. ； The effective coefficient is after environmental dynamic correction; The basic mapping parameters for offline calibration are: T is the real-time ambient temperature; H is the real-time relative humidity. Reference temperature; Reference humidity; This refers to the temperature sensitivity coefficient. This is the humidity sensitivity coefficient.

4. The intelligent control system for the switchgear according to claim 3, characterized in that: The specific calculation method for the arc energy assessment value during the circuit breaker opening process is as follows: In the offline phase based on nonlinear mapping relationships, tripping waveform data of different voltage levels and arc-extinguishing dielectric cartridges are acquired through a high-voltage breaking test bench. In the online phase, the moment of current zero-crossing is identified. Based on the discrete sampled data in the neighborhood at that moment, the absolute value of the rate of change of current is calculated using a numerical differentiation algorithm. Based on the real-time travel data measured by the displacement sensor, combined with the current zero-crossing time The timestamp is used to obtain the contact opening distance at that moment through an interpolation algorithm. The maximum instantaneous absolute value of the main circuit current during this tripping operation is obtained as the peak current of the main circuit. The arc energy assessment value is obtained based on the product of the absolute value of the current change rate, the contact gap, and the peak current of the main circuit. This arc energy assessment value is then labeled as... .

5. The intelligent control system for the switchgear according to claim 4, characterized in that: The adaptive control decision module specifically calculates the optimal tripping speed reference curve as follows: First, a comprehensive cost function is constructed, which includes an arc energy penalty term, a velocity tracking error term, and a control variable change rate penalty term. This comprehensive cost function is then labeled as... ; Subsequently, the model predictive control algorithm, within each fixed microsecond-level control cycle, based on the aforementioned comprehensive cost function... A constrained online rolling optimization solution is performed, and the solution process is subject to the following constraints: peak value constraint of coil drive current, maximum allowable acceleration constraint of contacts, and upper and lower limits constraint of duty cycle of power drive signal.

6. The intelligent control system for the switchgear according to claim 5, characterized in that: The model predictive control algorithm performs the following steps within each fixed microsecond-level control cycle: S51: Based on the predictive model of the operating mechanism, predict the displacement trajectory, velocity trajectory and corresponding arc energy assessment value of the contact under different candidate driving parameters in the future control cycle. S52: Under the condition that the above constraints are satisfied at the same time, the optimal control sequence is obtained by minimizing the preset comprehensive cost function through an online quadratic programming solver; S53: Output the first element of the optimal control sequence as the coil drive control parameter for the current control cycle; S54: When the next control cycle arrives, the prediction model is corrected by using the real-time feedback data collected by the multi-source sensing module, and steps S51 to S53 are repeated to achieve rolling optimization closed-loop control. The prediction model is a discrete-time state-space model of the operating mechanism, whose state variables include coil current, contact movement speed and contact displacement, and the control variable is the duty cycle of the PWM drive signal. The model is obtained by derivation based on physical mechanisms or system identification methods based on input and output data, and the model parameters are corrected online using real-time feedback data during operation.

7. The intelligent control system for the switchgear according to claim 6, characterized in that: The adaptive control decision module evaluates the effectiveness of the solution results of the model predictive control algorithm in each control cycle. The evaluation criteria include: (1) When the online quadratic programming solver returns an infeasible state, it is determined that the solution result exceeds the preset constraint boundary; (2) If the optimal control quantity calculated within three consecutive control cycles touches the upper or lower limit of the constraint boundary, and the arc energy assessment value does not show a downward trend, it is determined that the solution result has not converged within a consecutive preset number of cycles. When any of the above judgment conditions are met, the adaptive control decision module triggers a preset safety protection control strategy.

8. The intelligent control system for the switchgear according to claim 7, characterized in that: The safety margin control strategy includes: First control action: Stop using the optimal tripping speed reference curve generated in real time by the model predictive control algorithm, and instead load the reference tripping speed curve pre-stored in the non-volatile memory; the reference tripping speed curve is obtained by fitting the opening and closing data of this type of switchgear under various specified operating conditions in type tests; The second control action is to forcibly switch the control mode from closed-loop feedback adjustment mode to open-loop fixed duty cycle drive mode, and output drive signals according to the preset time-duty cycle table corresponding to the reference opening speed curve until the opening position signal is detected.

9. The intelligent control system for the switchgear according to claim 8, characterized in that: The calibration-free degradation compensation module specifically includes: Feature extraction unit: used to extract action feature parameters, including action time, average stroke speed, stroke end overshoot, and coil current peak value, from the coil current waveform and the actual contact stroke curve of each opening and closing operation; Degradation trend calculation unit: It is used to process the numerical sequence of action feature parameters of the same type in multiple consecutive operations using the exponential weighted moving average algorithm to calculate the degradation trend index after the current operation. The deviation of the feature parameters of the recent operation from the reference benchmark value is given the first weight, and the historical cumulative degradation trend is given the second weight. Compensation amount generation unit: used to input the calculated degradation trend index into the pre-generated mapping relationship between degradation degree and control parameter compensation amount, to obtain the control parameter compensation amount required for the current operating cycle; The mapping relationship between the degree of degradation and the compensation amount of the control parameters is established in advance in the following way: Accelerated aging tests were conducted on prototype switchgear of the same model. During the test, the changes in its action characteristic parameters relative to the initial state were measured periodically, and the increment of control parameters required to restore the action characteristic parameters to near the initial state was determined simultaneously. The mapping relationship was generated by fitting multiple sets of changes with corresponding increment data.

10. An intelligent control method for switchgear, wherein the method employs the intelligent control system for switchgear as described in any one of claims 1-9, characterized in that: Includes the following steps: S1: Acquire contact current signal, contact voltage signal and contact displacement signal, and perform filtering preprocessing; S2: Identify the moment when the current crosses zero, extract the current change rate, the contact opening distance, and the peak current of the main circuit at that moment, and combine them with the nonlinear mapping relationship after dynamic correction by ambient temperature and humidity to calculate the arc energy assessment value during the opening process in real time. S3: Using the calculated arc energy assessment value as the feedback quantity and minimizing the arc energy as the control objective, within the preset speed-stroke constraint space, a model predictive control algorithm is used to perform rolling optimization iterative solution to generate the optimal opening speed reference curve and the corresponding initial coil drive control parameters. S4: Based on the sequence of action characteristic parameters of each opening and closing operation, calculate the performance degradation trend index of the operating mechanism using the exponential weighted moving average algorithm, and generate the control parameter compensation amount based on the pre-stored mapping relationship between the degree of degradation and the control parameter compensation amount. S5: The control parameter compensation amount is superimposed on the initial coil drive control parameters to generate the final drive command to control the operating mechanism to perform the opening operation; the coil current waveform and actual contact stroke curve of this opening operation are collected to update the degradation trend index and serve as the basis for the next control calculation.

Images