System and method for predictive maintenance function of switch cabinet based on thermodynamics

Abstract

The invention relates to a thermodynamics-based system for a predictive maintenance function of a switch cabinet, and the system comprises a sensor unit which comprises a temperature measurement sensor and a current sensor and is used for collecting a temperature signal, a current signal and an environment temperature signal of a connection point of a main loop of the switch cabinet in real time; the alarm unit is used for sending out an alarm signal after a fault is predicted to occur when the switch cabinet is in non-full-load operation; and a control unit configured to: calculate a difference between the temperature collected from the sensor unit and an ambient temperature to obtain a temperature rise value; initializing a model based on the corresponding time, the collected current and the calculated temperature rise value to obtain five fitted parameters; performing temperature rise value prediction based on the collected current signals in the prediction time period by using the obtained five parameters; and when the switch cabinet is judged to have a fault according to the comparison result of the predicted temperature rise value and the collected measured value, the alarm unit is controlled to send out an alarm signal.

Description

Technical Field

[0001] The present disclosure relates to a system and method for predictive maintenance of switch cabinets, and more particularly to a system and method for predictive maintenance of switch cabinets based on thermodynamics. Background Art

[0002] In industry, equipment such as switch cabinets that are often found at customer sites generally do not operate at full load. The temperature rise alarm system will only alarm when the temperature exceeds 125°C (maximum temperature value). Usually, when the equipment fails (such as loose bolts), no alarm will be issued during normal operation, but if it is suddenly fully loaded, the equipment will immediately crash. Therefore, a system and method for predictive maintenance of switch cabinets is needed. At present, the relevant research on the health status and temperature rise prediction of switch cabinets involves methods such as machine learning, normalization, and temperature analysis models based on the distribution of temperature measurement points. Among them, most methods include those designed based on a time parameter, or only applicable to a section of current, or the prediction and alarm only rely on the predicted value, resulting in low accuracy of the prediction results, data adaptability, and model reliability, and false alarms are prone to occur. In addition, the existing research on the switch cabinet temperature rise predictive maintenance system and method is mostly composed of two parts: fitting and prediction. For successful fitting, it is necessary to be based on the equipment being in a healthy state during the fitting stage. Based on this, the following problems still exist: it is difficult to identify whether the equipment in the fitting stage is in a healthy state; if the equipment has failed in the fitting stage, the system will use the failed state as a benchmark, so the failure will never be identified, let alone an alarm signal will be issued in time. Summary of the invention

[0003] The present disclosure relates to a system and method for predictive maintenance function of switch cabinet based on thermodynamics.

[0004] According to one aspect of the present disclosure, a system based on thermodynamics for predictive maintenance of switch cabinets is provided, comprising: a sensor unit, comprising a temperature sensor and a current sensor, for real-time acquisition of temperature signals, current signals and ambient temperature signals of the main circuit connection points of the switch cabinet; an alarm unit, for issuing an alarm signal after predicting that a fault will occur when the switch cabinet is not operating at full load; and a control unit, configured to: calculate the difference between the temperature collected from the sensor unit and the ambient temperature to obtain a temperature rise value; initialize a model based on the corresponding time, the collected current and the calculated temperature rise value to obtain five fitted parameters; use the obtained five parameters to perform temperature rise value prediction based on the collected current signal within the prediction time period; and when the switch cabinet is judged to be faulty based on the comparison result of the predicted temperature rise value and the collected measured value, control the alarm unit to issue an alarm signal.

[0005] Optionally, the system further includes: wherein, the judging that the switch cabinet is faulty includes that consecutive comparison results are all higher than an upper limit 5k of a 95% confidence interval of the predicted temperature rise value.

[0006] Optionally, the system also includes: wherein the five parameters are five parameters for a dual time precision five-parameter fitting algorithm, including a first time constant, a second time constant, a parameter representing a proportion of the first time constant or the second time constant, a parameter representing a temperature jump from a first current to a second current and from a first temperature to a second temperature, and a maximum value of the temperature rise value.

[0007] Optionally, the system further includes: wherein the sensor unit starts collecting data at any time.

[0008] Optionally, the system further comprises: wherein the sensor unit collects 8 hours of data from the start of data collection for the initialization.

[0009] Optionally, the system further comprises: wherein the control unit performs a 5% outlier noise removal operation on the collected current signal.

[0010] Optionally, the system further comprises: when the control unit detects that the number of current segments collected is less than 3, using five default parameters, performing temperature rise value prediction based on the current signal collected within the prediction time period.

[0011] Optionally, the system further includes: wherein the corresponding time includes the time carried by the collected signal or the time when the control unit receives the collected signal.

[0012] Optionally, the system further comprises: after the model is initialized, using the obtained five parameters or the default five parameters to perform temperature rise value prediction based on the full-load current signal to self-check whether the switch cabinet is in a healthy state.

[0013] Optionally, the system further includes: wherein, when the predicted temperature rise value is greater than a temperature rise value threshold, the alarm unit sends an alarm signal.

[0014] Optionally, the system further includes: wherein the temperature rise value threshold includes 125°C.

[0015] According to another aspect of the present disclosure, there is provided a method for using a system based on thermodynamics for predictive maintenance function of switch cabinets, the system comprising a sensor unit including a temperature sensor and a current sensor, an alarm unit and a control unit, the method comprising: the sensor unit collecting temperature signals, current signals and ambient temperature signals of the main circuit connection points of the switch cabinet in real time; the control unit calculating the difference between the temperature collected from the sensor unit and the ambient temperature to obtain a temperature rise value; the control unit initializing a model based on the corresponding time, the collected current and the calculated temperature rise value to obtain fitted five parameters; the control unit using the obtained five parameters to perform temperature rise value prediction based on the collected current signal within the prediction time period; and when the switch cabinet is judged to be faulty according to the comparison result of the predicted temperature rise value and the collected actual measured value, the control unit controlling the alarm unit to send an alarm signal.

[0016] Optionally, the method also includes: wherein the five parameters are five parameters for a dual time precision five-parameter fitting algorithm, including a first time constant, a second time constant, a parameter representing a proportion of the first time constant or the second time constant, a parameter representing a temperature jump from a first current to a second current and from a first temperature to a second temperature, and a maximum value of the temperature rise value.

[0017] Optionally, the method further includes: after the model is initialized, the control unit uses the obtained five parameters or the default five parameters to perform temperature rise value prediction based on the full-load current signal to self-check whether the switch cabinet is in a healthy state.

[0018] Optionally, the method further includes: wherein, judging that the switch cabinet is faulty includes that consecutive comparison results are all higher than an upper limit 5k of a 95% confidence interval of the predicted temperature rise value.

[0019] Optionally, the method further includes: wherein the sensor unit starts collecting data at any time.

[0020] Optionally, the method further comprises: wherein the sensor unit collects 8 hours of data from the start of data collection for the initialization.

[0021] Optionally, the method further includes: wherein the control unit performs a 5% outlier noise removal operation on the collected current signal.

[0022] Optionally, the method further includes: when the control unit detects that the number of current segments collected is less than 3, using default five parameters, performing temperature rise value prediction based on the current signal collected within the prediction time period.

[0023] Optionally, the method further includes: wherein the corresponding time includes the time carried by the collected signal or the time when the control unit receives the collected signal.

[0024] Optionally, the method further includes: when the predicted temperature rise value is greater than a temperature rise value threshold, the alarm unit sends an alarm signal.

[0025] Optionally, the method further includes: wherein the temperature rise value threshold includes 125°C.

[0026] Beneficial Effects

[0027] The present invention utilizes the thermal properties of the conductor and uses multivariate nonlinear regression to implement a method of fitting five parameters for a temperature rise prediction model, thereby realizing the function of real-time monitoring of the temperature rise of key connection points of the equipment under variable current conditions. An alarm signal can be issued in real time when a failure occurs during equipment operation at less than full load, thereby solving the limitation of the prior art of using fixed thresholds for alarms. On the premise of knowing the current signal, the future temperature rise can be predicted.

[0028] In addition, the present invention does not require heating simulation and does not require the construction of any model. The five parameters are all obtained based on thermodynamic parameters and are related to the material properties of the measuring points. Therefore, the method has good portability and is suitable for temperature rise prediction and fault alarm of the main circuit connection points of all switch cabinets.

[0029] The system of the present invention runs fast and simply. When the working conditions remain unchanged, only one parameter fitting is required. The algorithm is very light in magnitude after encapsulation (C environment), and only tens of milliseconds are required for one run, which greatly improves the calculation efficiency. Since the present invention uses a dual-time precision five-parameter fitting algorithm, the prediction accuracy is also improved.

[0030] In addition, the switch cabinet health status self-check method proposed in the present invention can identify whether the fitting stage of the switch cabinet predictive maintenance system is carried out under the healthy state of the equipment, thereby ensuring that the data in the prediction stage is correct and reliable, making the evaluation of the health status of the switch cabinet more intelligent; at the same time, the solution also provides a new method for identifying equipment failures under non-full load conditions, thereby improving the reliability of the system and achieving the purpose of reducing customer losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The aspects, features and advantages of the present disclosure will become clearer and easier to understand through the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0032] Figure 1 A unit diagram of a system for predictive maintenance of switch cabinets based on thermodynamics according to an embodiment of the present disclosure is shown;

[0033] Figure 2A flow chart showing a method for predictive maintenance of a switch cabinet based on thermodynamics according to an embodiment of the present disclosure;

[0034] Figure 3 A data acquisition curve diagram of a system for predictive maintenance of switch cabinets based on thermodynamics according to an embodiment of the present disclosure is shown;

[0035] Figure 4 A comparison diagram of a fault temperature rise curve and a predicted temperature rise curve of a system for predictive maintenance of a switch cabinet based on thermodynamics according to an embodiment of the present disclosure is shown;

[0036] Figure 5 A schematic diagram showing the results of a method for health self-checking of a system for predictive maintenance of a switch cabinet based on thermodynamics according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0037] Before proceeding to the following detailed description, it may be advantageous to set forth the definitions of certain words and phrases used throughout this patent document. The terms "coupling", "connection" and their derivatives refer to any direct or indirect communication or connection between two or more elements, regardless of whether those elements are in physical contact with each other. The terms "transmission", "reception" and "communication" and their derivatives cover direct and indirect communication. The terms "include" and "comprise" and their derivatives refer to including but not limited to. The term "or" is inclusive, meaning and / or. The phrases "associated with...", "corresponding to..." and their derivatives refer to including, including within, interconnecting, including, including within, connecting or connecting with, coupling or coupling with, communicating with, cooperating, interweaving, parallel, approaching, binding or binding with, having, having attributes, having a relationship or having a relationship with, etc. The term "controller" refers to any device, system or part thereof that controls at least one operation. Such a controller can be implemented with hardware, or a combination of hardware and software and / or firmware. The functions associated with any particular controller can be centralized or distributed, whether local or remote. The phrase "at least one of", when used with a list of items, means that different combinations of one or more of the listed items may be used, and only one of the items in the list may be required. For example, "at least one of A, B, C" includes any of the following combinations: A, B, C, A and B, A and C, B and C, A and B and C.

[0038] Definitions for other specific words and phrases are provided throughout this patent document. Those of ordinary skill in the art should understand that in many, if not most instances, such definitions apply to prior and future uses of such defined words and phrases.

[0039] In this patent document, the application combination of units and the division level of subunits are only used for illustration. Without departing from the scope of the present disclosure, the application combination of units and the division level of subunits can have different modes. The embodiments of the present disclosure can be implemented in different forms and should not be interpreted as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to make the present disclosure thorough and complete, and to fully convey exemplary implementation methods to those skilled in the art. The embodiments of the present disclosure can be combined to form additional embodiments.

[0040] The present disclosure will be described in detail below with reference to exemplary embodiments of the present disclosure. However, the present disclosure is not limited to the embodiments described herein, and it can be implemented in many different forms. The described embodiments are only used to make the present disclosure thorough and complete, and to fully convey the concept of the present disclosure to those skilled in the art. The features of the various embodiments described can be combined or replaced with each other, unless explicitly excluded or should be excluded according to the context.

[0041] Figure 1 A unit diagram of a system for predictive maintenance of switch cabinets based on thermodynamics according to an embodiment of the present disclosure is shown.

[0042] like Figure 1 As shown, the system 100 according to the embodiment of the present disclosure may include a sensor unit 101, an alarm unit 102, and a control unit 103. The sensor unit 101 includes a temperature sensor 1011 and a current sensor 1012. In some embodiments, the temperature sensor 1011 can collect temperature signals and ambient temperature signals of the main circuit connection points of the switch cabinet in real time. In some embodiments, the collected temperature signals and ambient temperature signals can carry a time series containing time by themselves. Optionally, in some embodiments, the control unit 103 can process the signals immediately upon receiving the collected temperature signals and ambient temperature signals, and obtain the corresponding time of the signals. In some embodiments, the current sensor 1012 can collect current signals of the main circuit connection points of the switch cabinet in real time. In some embodiments, the collected current signals can carry a time series containing time by themselves. Optionally, in some embodiments, the control unit 103 can process the signals immediately upon receiving the collected current signals, and obtain the corresponding time of the signals. In some embodiments, the sensor unit 101 will collect 8 hours of data from the time when the data collection starts for the initialization model operation of the control unit 103. In some embodiments, the control unit 103 may be configured to perform a 5% outlier noise removal operation on the collected current signal.

[0043] In some embodiments, the sensor unit 101 may transmit the collected temperature signal, current signal and ambient temperature signal to the control unit for further calculation and processing. In some embodiments, the control unit 103 may process the collected signal immediately upon receiving the collected signal and obtain the corresponding time of the signal. Optionally, in some embodiments, the collected signal may carry a time series containing time, which is transmitted to the control unit 103 as the corresponding time. In some embodiments, the control unit 103 may be configured to calculate the difference between the temperature collected from the sensor unit and the ambient temperature to obtain the temperature rise value. In some embodiments, the control unit 103 may be configured to initialize the model based on the corresponding time, the current collected by the sensor unit and the calculated temperature rise value to obtain the fitted five parameters. In some embodiments, the control unit 103 may be configured to use the obtained five parameters to perform temperature rise value prediction based on the collected current signal within the prediction time period. In some embodiments, the control unit 103 may be configured to control the alarm unit to send an alarm signal when the switch cabinet is judged to be faulty according to the comparison result of the predicted temperature rise value and the collected measured value. In some embodiments, the control unit 103 may be configured to determine that the continuous comparison results are all higher than the upper limit 5k of the 95% confidence interval of the predicted temperature rise value as a switch cabinet failure.

[0044] In some embodiments, the sensor unit 101 according to the embodiment of the present disclosure can collect temperature signals, current signals and ambient temperature signals for multiple current segments in real time. Therefore, in some embodiments, the control unit 103 can be configured to use the default five parameters when the number of segments of the current collected by the sensor unit is less than 3, without initializing the model to obtain the fitted five parameters. Further, in some embodiments, the control unit 103 can be configured to use the default five parameters to perform temperature rise value prediction based on the collected current signal within the prediction time period. In some embodiments, the control unit 103 can be configured to control the alarm unit to send an alarm signal when the switch cabinet is judged to be faulty according to the comparison result of the predicted temperature rise value and the collected measured value. In some embodiments, the control unit 103 can be configured to judge that the continuous comparison results are all higher than the upper limit 5k of the 95% confidence interval of the predicted temperature rise value as a switch cabinet fault. Among them, the number of continuous comparison results can be 20 continuously sampled data, but is not limited to this. The number and the condition for judging as a fault can be reset.

[0045] In some embodiments, the control unit 103 according to the embodiment of the present disclosure may be configured to perform a temperature rise prediction based on a full-load current signal after completing the initialization model to self-check whether the switch cabinet is in a healthy state. In some embodiments, the control unit 103 may be configured to determine that the switch cabinet state itself is in a fault state when the predicted temperature rise value is greater than the temperature rise value threshold. The temperature rise value threshold may include 125°C and 140°C, but is not limited thereto, and the temperature rise value threshold may be reset according to demand, the material's extreme temperature threshold, etc.

[0046] Figure 2 A flow chart of a method for predictive maintenance of a switch cabinet based on thermodynamics according to an embodiment of the present disclosure is shown.

[0047] like Figure 2 As shown, the method according to the embodiment of the present disclosure can be executed according to steps S200-S209 to obtain five parameters. The five parameters in the present disclosure are all obtained based on thermodynamic parameters and are related to the material properties of the measuring point. In step S200, the sensor unit 101 can collect data in real time to obtain time, current, and temperature rise value. The current signal, temperature signal, and ambient temperature signal previously collected in real time can be transmitted to the control unit 103. The control unit 103 can calculate the difference based on the temperature signal collected by the sensor unit and the ambient temperature signal to obtain the temperature rise value. The corresponding time, the collected current, and the temperature rise value can be used for further processing. Among them, the sensor unit 101 includes a temperature sensor 1011 and a current sensor 1012. The temperature sensor 1011 can collect the temperature signal and the ambient temperature signal of the main circuit connection point of the switch cabinet in real time, and the current sensor 1012 can collect the current signal of the main circuit connection point of the switch cabinet in real time. The corresponding time can include the collected temperature signal, the ambient temperature signal, and the time series containing time carried by the current signal itself. Optionally, the corresponding time can include the time when the control unit 103 receives the collected signal. According to the thermal properties of resistance and the fact that the current temperature rise is closely related to the previous state, data collection can start at any time. The sensor unit 101 needs to collect 8 hours (8h) of data to help successfully complete the subsequent initialization operation to obtain the fitted five parameters.

[0048] In step S201 , during the process of scrolling and reading data, the control unit 103 may intercept data.

[0049] In step S202 , the control unit 103 may perform a 5% outlier noise removal operation on the collected current signal, so as to facilitate subsequent accurate determination of the current segment number and reduce the impact of current fluctuations on the algorithm.

[0050] Only when the current in the initial stage changes twice or more can the parameter attributes suitable for initialization be captured, that is, the fitting of beta (β) in the five parameters can be completed. In step S203, the control unit determines whether the number of current segments is less than 3. If it is calculated that the number of current segments in the initialization stage is less than 3, the five parameters can be directly assigned as the default five parameters in step 204. Because, at the customer site, the collected data may include constant changes and inconvenient situations, and some loads may not change in current or temperature during the time period of data collection, so it may become difficult to initialize and obtain the five parameters, so it is possible to directly use the typical default five parameters of Schneider products for subsequent predictions in step S208 and fault alarms in step S209. The default five parameters are estimated based on the typical cabinet type of the product. Compared with the five parameters obtained by calculation, the predicted temperature rise value of the estimated five parameters will lose some accuracy, but it can quickly complete the algorithm initialization and output relatively reliable prediction values ​​without the initialization conditions. In addition, different typical default five-parameter values ​​can be provided accordingly according to the different products used in different customer scenarios and the changes in the installation locations of the measuring points.

[0051] If it is calculated that the number of current segments in the initialization phase is not less than 3, the control unit can determine whether the initialization is completed in step 205. If the initialization is not completed, it can return to the data interception step (S102) to re-select new initialization data.

[0052] If the initialization has been completed, the switch cabinet health self-check can be performed in step 206 to determine whether the equipment is healthy, that is, the health self-check of the model. Because if the equipment is already in a faulty state during initialization, the wrong five parameters can still be fitted. If the wrong five parameters are used as a benchmark for subsequent predictions and alarms, wrong prediction results and false alarms will be obtained, which seriously affects the reliability of the model. Therefore, the health self-check of the model is necessary and must be done.

[0053] Five parameters are obtained by fitting the initialization state data. In this operating state, the full load current of 4h is directly input. Whether the initial state is faulty is determined by whether the maximum predicted temperature rise value under full load current exceeds the temperature threshold. The temperature threshold includes 125℃ or 140℃, but is not limited to this. It can be reset according to the demand, the limit temperature threshold of the material, etc. When the initialization state is healthy, the maximum predicted temperature rise of the maximum current for 4h will not exceed the temperature threshold such as 125℃ when it is stable; when the initialization state is faulty, the predicted temperature rise of the maximum current for 4h will exceed the temperature threshold such as 125℃ before it reaches stability. The initialization self-check achieved in this way can ensure that the initialization is carried out under the benchmark of the healthy state of the equipment and model, making the evaluation of the health state of the switch cabinet more intelligent.

[0054] At the same time, if the health self-check status feedback of the model is No, it means that the equipment is already in a faulty state. In step S209, the control unit can control the alarm unit to send an alarm signal and execute the fault alarm function to prompt the customer to repair the equipment in time, which can further reduce customer losses. According to an embodiment of the present disclosure, a new method for identifying equipment failures when the equipment is not fully loaded is proposed, making the health status assessment of the switch cabinet more diversified. In addition, in some embodiments, the health self-check assessment process can also provide customers with an alarm time when the temperature threshold can be reached.

[0055] If the initialization device is healthy (i.e., the feedback is Yes), then in step S207, the five fitting parameters can be obtained. The method includes a dual time precision five-parameter fitting algorithm. The five parameters include α, β, τ1, τ2 and θ i_max (Trmax). Among them, the average value I calculated by multi-segment current R Stable temperature rise value θ i (t) The prediction formula is as follows:

[0056]

[0057] Among them, θ i_max Can represent the highest value of temperature rise, α i and 1-α i Can be used to represent the first time constant τ around the measuring point i1 and the first time constant τ i2 The proportion of i It can represent a jump from a first current to a second current or a jump from a first temperature to a second temperature. t can represent time, and i represents the actual current to be predicted.

[0058] In step S208, the control unit can use the obtained five parameters or the default five parameters to perform temperature rise value prediction based on the current signal collected within the prediction time period. When the switch cabinet is judged to be faulty according to the comparison result between the predicted temperature rise value and the collected measured value, in step S209, the control unit can control the alarm unit to send an alarm signal. Judging that the switch cabinet is faulty includes that the continuous comparison results are all higher than the upper limit 5k of the 95% confidence interval of the predicted temperature rise value. Among them, the number of continuous comparison results can be 20 continuously sampled data, but is not limited to this, and can be reset according to demand, that is, 20 data can be continuously sampled before making a fault judgment and determining whether to issue an alarm. And in order to increase the credibility of the results of the method, provide more information about the predicted value, and avoid the occurrence of false alarms, the above-mentioned 95% confidence interval is added to help improve the reliability of the prediction results.

[0059] Figure 3 FIG. 1 shows a data acquisition curve diagram of a system for predictive maintenance of switch cabinets based on thermodynamics according to an embodiment of the present disclosure. Figure 3 As shown, the initialization data can be captured at any time. From the start of data collection, the sensor unit will collect 8 hours of data to complete the initialization. Among them, the first 4 hours of data only provide the correlation information between the previous and next data, and the last 4 hours are actually involved in the five-parameter fitting. Therefore, 8 hours of data are ultimately required to complete the five-parameter initialization.

[0060] Figure 4 A comparison diagram of the fault temperature rise curve and the predicted temperature rise curve of the system for predictive maintenance of switch cabinets based on thermodynamics according to an embodiment of the present disclosure is shown. Figure 4 As shown, when a device fails, the system according to the embodiment of the present disclosure can identify the change in the temperature rise of the device in real time and send out an alarm signal. In this embodiment, a comparison experiment between the healthy state and the fault state of the device is designed. The fault experiment is to loosen the B-phase bolts and compare the measured value with the predicted value in real time. If 20 consecutive sampling data are higher than the predicted value of the 95% confidence upper limit of the algorithm, a predictive maintenance alarm is given. Figure 4 In the example, the prediction curve is close to the healthy state experimental data, while the fault state experimental data continues to be more than 5K away from the prediction upper line, at which time the alarm function will be triggered. This shows that the system can identify the fault state of the equipment and issue an alarm signal.

[0061] Figure 5 The result diagram of the method for self-checking the health of a system for predictive maintenance of a switch cabinet based on thermodynamics according to an embodiment of the present disclosure is shown. Figure 5 As shown, once the initialization is completed, the five fitted parameters can be determined. When the predicted current value of the full load is input, the control unit can use the determined five parameters to output the predicted temperature rise value corresponding to the full load current. If the switch cabinet is already in a faulty state, the fitted curve is far different from the measured value at this time. The determined five parameters are the five parameters in the faulty state, and the predicted temperature rise value can no longer be used as a reference value to determine whether the equipment is faulty. In some embodiments, the equipment (switch cabinet) that has been judged to be faulty can send an alarm through the alarm unit. In addition, in some embodiments, the health self-test assessment process can also provide customers with an alarm time when the temperature threshold can be reached.

[0062] The hardware computing device described in the present disclosure may be implemented in its entirety or in parts thereof through various suitable hardware means, including but not limited to FPGA, ASIC, SoC, discrete gate or transistor logic, discrete hardware components, or any combination thereof.

[0063] The block diagrams of circuits, devices, apparatuses, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As will be appreciated by those skilled in the art, these circuits, devices, apparatuses, equipment, and systems may be connected, arranged, or configured in any manner as long as the desired purpose can be achieved.

[0064] Those skilled in the art should understand that the above-mentioned specific embodiments are merely examples and not limitations, and various modifications, combinations, partial combinations and replacements may be made to the embodiments of the present disclosure according to design requirements and other factors. As long as they are within the scope of the attached claims or their equivalents, they belong to the scope of rights to be protected by the present disclosure.

Claims

1. A system for predictive maintenance of switch cabinet based on thermodynamics, comprising: The sensor unit includes a temperature sensor and a current sensor, which are used to collect temperature signals, current signals and ambient temperature signals of the main circuit connection points of the switch cabinet in real time; An alarm unit is used to send out an alarm signal after predicting a fault when the switchgear is not operating at full load; and A control unit configured to: Calculating the difference between the temperature collected from the sensor unit and the ambient temperature to obtain a temperature rise value; Initializing the model based on the corresponding time, the collected current and the calculated temperature rise value to obtain the fitted five parameters; Using the obtained five parameters, a temperature rise value prediction is performed based on the current signal collected within the prediction time period; as well as When the switch cabinet is judged to be faulty according to the comparison result between the predicted temperature rise value and the collected measured value, the alarm unit is controlled to send out an alarm signal.

2. The system of claim 1, wherein: The determining that the switch cabinet is faulty includes that the continuous comparison results are all higher than the upper limit 5K of the 95% confidence interval of the predicted temperature rise value.

3. The system of claim 1, wherein: The five parameters are five parameters used for the dual time precision five-parameter fitting algorithm, including the first time constant, the second time constant, a parameter indicating the proportion of the first time constant or the second time constant, a parameter indicating the temperature jumps from the first current to the second current and from the first temperature to the second temperature, and the maximum value of the temperature rise value.

4. The system of claim 1, wherein: The sensor unit starts collecting data at any time.

5. The system of claim 4, wherein: The sensor unit will collect data for 8 hours from the time when data collection starts for the initialization.

6. The system of claim 1, wherein: The control unit performs a 5% outlier noise removal operation on the collected current signal.

7. The system according to claim 1, further comprising: when the control unit detects that the number of current segments collected is less than 3, using default five parameters, performing temperature rise value prediction based on the current signal collected within the prediction time period.

8. The system of claim 1, wherein: The corresponding time includes the time carried by the collected signal or the time when the control unit receives the collected signal.

9. The system according to any one of claims 1 to 8, further comprising, after the model is initialized, using the obtained five parameters or the default five parameters to perform temperature rise value prediction based on the full-load current signal to self-check whether the switch cabinet is in a healthy state.

10. The system of claim 9, wherein: When the predicted temperature rise value is greater than the temperature rise value threshold, the alarm unit sends out an alarm signal.

11. The system of claim 10, wherein: The temperature rise threshold value includes 125°C.

12. A method for using a system based on thermodynamics for predictive maintenance of a switch cabinet, the system comprising a sensor unit including a temperature sensor and a current sensor, an alarm unit and a control unit, the method comprising: The sensor unit collects temperature signals, current signals and ambient temperature signals of the main circuit connection points of the switch cabinet in real time; The control unit calculates the difference between the temperature collected from the sensor unit and the ambient temperature to obtain a temperature rise value; The control unit initializes the model based on the corresponding time, the collected current and the calculated temperature rise value to obtain the fitted five parameters; The control unit uses the obtained five parameters to perform temperature rise value prediction based on the current signal collected within the prediction time period; as well as When the switch cabinet is judged to be faulty according to the comparison result between the predicted temperature rise value and the collected measured value, the control unit controls the alarm unit to send out an alarm signal.

13. The method of claim 12, wherein: The five parameters are five parameters used for the dual time precision five-parameter fitting algorithm, including the first time constant, the second time constant, a parameter indicating the proportion of the first time constant or the second time constant, a parameter indicating the temperature jumps from the first current to the second current and from the first temperature to the second temperature, and the maximum value of the temperature rise value.

14. The method according to any one of claims 12 to 13 further comprises, after the model is initialized, the control unit using the obtained five parameters or the default five parameters to perform a temperature rise prediction based on the full-load current signal to self-check whether the switch cabinet is in a healthy state.

Images