A locomotive toilet antifreeze emptying intelligent control method and system

Abstract

The present application relates to the technical field of antifreeze for locomotive sanitary facilities, and specifically to a locomotive toilet antifreeze emptying intelligent control method and system, the method comprising: using a dynamic weighted fusion algorithm to dynamically allocate weights for multi-source locomotive toilet-related data, quantifying the dynamic risk index of multi-source locomotive toilet-related data within the current analysis period; combining the long-term risk trends and short-term risk mutations of all dynamic risk indices calculated for all collected analysis periods, and fusing the two to construct a long-term and short-term dual-modal icing risk index; and using the dual-modal icing risk index to set a multi-level threshold antifreeze mechanism strategy. The present application aims to enable locomotive toilets to dynamically adapt to extreme cold environments, reduce energy consumption, and avoid misjudgments and failures caused by static weight allocation and single indicator evaluation in traditional systems through dynamic long-term and short-term risk analysis and graded antifreeze processing of multi-source locomotive toilet-related data.

Description

Technical Field

[0001] The present application relates to the technical field of antifreeze of locomotive sanitary facilities, and in particular to an intelligent control method and system for antifreeze emptying of locomotive toilets. Background Art

[0002] Traditional locomotive sanitation systems often malfunction in low-temperature environments due to problems such as sewage retention and pipe freezing. This not only affects the normal operation of equipment, brings health hazards and environmental pollution, but also seriously threatens the locomotive's punctuality and operational safety.

[0003] In existing locomotive toilet antifreeze and emptying systems, traditional electric heating or constant temperature heating modes rely on continuous electrical energy input to maintain the temperature of pipes and liquid storage tanks. In extremely cold environments, these systems need to run at full power for a long time, resulting in a sharp increase in energy consumption. Specifically, traditional toilet antifreeze and emptying control systems trigger heating or emptying operations through preset temperature thresholds or fixed weight distributions, but ignore the dynamic coupling effects of locomotive toilet-related data such as pipe temperature, sewage level, locomotive vibration, and ambient humidity. This design easily leads to frequent misjudgments of the system in extremely cold environments. In order to compensate for this misjudgment defect, the toilet antifreeze and emptying control system is forced to adopt a continuous full-power heating mode, which not only increases energy consumption, but also easily induces microcracks in the pipe material, exacerbating the infiltration of cold air and the risk of freezing, thus forming a vicious cycle of "high energy consumption-low reliability". Summary of the Invention

[0004] In order to solve the above technical problems, the present application provides a method and system for intelligent control of antifreeze emptying of locomotive toilets. The technical solutions adopted are as follows:

[0005] In a first aspect, an embodiment of the present application provides an intelligent control method for anti-freezing and emptying of a locomotive toilet, the method comprising the following steps:

[0006] Collecting and preprocessing multi-source locomotive toilet-related data in the locomotive toilet system; wherein the multi-source locomotive toilet-related data is collected by multiple sensors installed in pipes, sewage storage tanks, locomotive chassis and the outside of the locomotive body;

[0007] Use a dynamic weighted fusion algorithm to dynamically assign weights to multi-source locomotive toilet-related data and quantify the dynamic risk index of multi-source locomotive toilet-related data within the current analysis period;

[0008] Combine the long-term risk trends and short-term risk mutations of all dynamic risk indices calculated during all analysis periods and fuse them to construct a long-term and short-term dual-modal icing risk index;

[0009] A multi-level threshold antifreeze mechanism strategy is set using a dual-modal icing risk index.

[0010] Preferably, the multi-source locomotive toilet related data includes: pipeline temperature, sewage level, locomotive vibration and ambient humidity.

[0011] Preferably, the method of dynamically allocating weights of multi-source locomotive toilet related data using a dynamic weighted fusion algorithm and quantifying the dynamic risk index of multi-source locomotive toilet related data in the current analysis period is:

[0012] The entropy weight method and least squares method are used to determine the weight of each type of locomotive toilet related data and the slope of the fitting line in the current analysis period respectively. The result of weighted fusion of the weights of all types of locomotive toilet related data and the absolute value of the slope of the fitting line is used as the dynamic risk index of multi-source locomotive toilet related data in the current analysis period.

[0013] Preferably, the weighted fusion is an operation of performing a weighted summation of the absolute values ​​of the slopes of the fitted straight lines with respect to the weights of all types of data.

[0014] Preferably, the long-term risk trend is further determined by extracting long-term trends from all dynamic risk indices using a sliding average algorithm.

[0015] Preferably, the short-term risk mutation is further determined by using a CUSUM cumulative control algorithm to detect short-term mutations of all dynamic risk indices.

[0016] Preferably, the method for constructing the long-term and short-term dual-modal icing risk index is further obtained by multiplying the long-term risk trend and the short-term risk mutation by a normalized result.

[0017] Preferably, the method of setting a multi-level threshold antifreeze mechanism strategy by utilizing the dual-mode icing risk includes:

[0018] Set multi-level thresholds to trigger anti-freeze measures at low, medium, and high risk levels based on the bimodal icing risk index;

[0019] Maintaining basic surveillance mode at low risk level;

[0020] Initiate a dynamic electric heat tracing strategy at a medium risk level;

[0021] Activates emergency drain procedure and switches to full power electric heat tracing mode at high risk levels.

[0022] Preferably, the method for setting a multi-level threshold and triggering low, medium and high risk antifreeze measures according to the dual-modal icing risk index is as follows:

[0023] Setting a first threshold and a second threshold; wherein the first threshold is less than the second threshold;

[0024] When the bimodal icing risk is less than the first threshold, the locomotive toilet is at a low risk level;

[0025] When the bimodal icing risk is greater than or equal to the first threshold and less than the second threshold, the locomotive toilet is at a medium risk level;

[0026] When the bimodal icing risk is greater than or equal to the second threshold, the locomotive toilet is at a high risk level.

[0027] In a second aspect, another embodiment of the present application further provides a locomotive toilet antifreeze emptying intelligent control system, which implements the above-mentioned locomotive toilet antifreeze emptying intelligent control method, the system comprising:

[0028] Multi-source locomotive toilet related data collection and preprocessing module, used to collect multi-source locomotive toilet related data in the locomotive toilet system and perform preprocessing;

[0029] A dynamic risk modeling module is used to dynamically assign weights to multi-source locomotive toilet-related data using a dynamic weighted fusion algorithm, and quantify the dynamic risk index of multi-source locomotive toilet-related data within the current analysis period;

[0030] The long-term and short-term risk fusion module is used to combine the long-term risk trend and short-term risk mutation of all dynamic risk indices calculated in all analysis periods, and fuse the two to construct a long-term and short-term dual-modal icing risk index;

[0031] The hierarchical antifreeze control module is used to set a multi-level threshold antifreeze mechanism strategy using a dual-modal icing risk index.

[0032] This application has at least the following beneficial effects:

[0033] (1) To address the problems that single sensor data cannot fully reflect the status of the locomotive toilet system and static weight allocation cannot adapt to the dynamic environment, a dynamic weighted fusion algorithm is used to dynamically allocate the weights of multi-source locomotive toilet-related data. The least squares method is combined to quantify the icing probability under the synergistic effect of multiple parameters to solve the problems of risk assessment bias and anti-freeze control failure.

[0034] (2) To address the problem of short-term fluctuations masking long-term hidden dangers or sudden mutations not being captured in a timely manner, a sliding average algorithm is used to extract long-term trends, combined with a CUSUM algorithm to detect short-term cumulative deviations, to balance the persistence and suddenness of risks, and to eliminate the interference of a single indicator's one-sidedness on the control strategy;

[0035] (3) Through the hierarchical threshold control strategy, the progressive responses such as electric heating and emergency emptying are dynamically triggered, and the design of heating the area with a larger slope of the locomotive toilet is given priority, which alleviates the local temperature difference and microcracks caused by the difference in thermal conductivity of the material and prolongs the life of the pipeline; the locomotive vibration data is combined to optimize the emptying timing, and mechanical vibration is used to assist the flow of sewage, thereby improving the emptying efficiency and reducing the energy consumption of the air pump; the edge computing unit realizes real-time data processing and decision-making closed loop, which improves the system response speed, ensures reliability and stability in extreme environments, and achieves the beneficial effects of reducing energy consumption, reducing operation and maintenance costs, and improving anti-freeze reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following is a brief introduction to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0037] Figure 1 A flowchart of an intelligent control method for anti-freezing and emptying a locomotive toilet provided in one embodiment of the present application. DETAILED DESCRIPTION

[0038] An embodiment of the present application provides an intelligent control method for antifreeze emptying of a motorcycle toilet. Figure 1 , the method comprises the following steps:

[0039] Step 1: Collect multi-source locomotive toilet related data in the locomotive toilet system and perform pre-processing.

[0040] In the locomotive toilet anti-freezing and emptying intelligent control system, in order to ensure the comprehensiveness and accuracy of data collection, high-precision temperature sensors are installed on the pipelines to monitor the temperature changes of the pipelines in real time; capacitive liquid level sensors are installed in the sewage storage tanks to dynamically detect the sewage inventory and judge the emptying needs; vibration sensors are deployed on the locomotive chassis and connections to synchronously collect the impact of the locomotive operating status on pipeline vibration; humidity sensors are installed on the outside of the car body to collect external environment humidity data to assess the risk of icing.

[0041] These sensors synchronously transmit multi-dimensional data at a 10Hz frequency to the edge computing unit, providing a foundation for subsequent analysis. A wavelet denoising algorithm eliminates high-frequency noise interference in the sensor signals to ensure data authenticity. A sliding window normalization method maps data of different dimensions to the [-1, 1] interval to eliminate the negative impact of magnitude differences on model training. The mapped data is then sorted in ascending chronological order.

[0042] The time period within a preset duration before each collection moment is used as the analysis period for the corresponding collection moment. In this embodiment, the preset duration is set to 1 hour. This application only intercepts the time series data within the analysis period of the current collection moment to construct the temperature sequence, liquid level sequence, vibration sequence, and humidity sequence at the current collection moment. The time period constraint within the preset duration ensures the timeliness and dynamic relevance of the sequence.

[0043] Step 2: Use the dynamic weighted fusion algorithm to dynamically assign weights to multi-source locomotive toilet related data and quantify the dynamic risk index of multi-source locomotive toilet related data within the current analysis period.

[0044] Because locomotives operate in extremely cold environments, sewage retention and pipe freezing are particularly prominent. The sewage pipes in locomotive toilets are prone to forming localized low-temperature retention points due to the locomotive's frequent starts and stops, as well as long periods of parking in cold areas. Traditional electric heating or constant-temperature heating methods rely on continuous high energy consumption to maintain pipe temperature, resulting in significant energy waste. For example, due to differences in material thermal conductivity, the sewage storage tank and pipes in locomotive toilets are prone to localized temperature differences during continuous heating, inducing microcracks at the pipe joints. Cold air infiltration through these cracks further exacerbates the risk of internal freezing.

[0045] Furthermore, locomotive toilet systems are significantly affected by dynamic factors such as locomotive vibration and sudden changes in external humidity. Traditional solutions rely on data from a single temperature sensor (e.g., monitoring only the temperature of the sewage tank), making it difficult to fully reflect the true status of each section of the pipeline, elbows, and drain valves. The static weight distribution mechanism cannot adapt to dynamic scenarios such as sewage sloshing caused by vibration and surface condensation caused by sudden increases in humidity, leading to risk assessment bias. For example, when a locomotive passes through a snowstorm, the humidity sensor data on the outside of the vehicle body changes dramatically, but the static weight fails to promptly prioritize the humidity parameter, causing the system to underestimate the risk of surface icing, ultimately causing the anti-freeze control to fail, threatening the normal emptying function of the locomotive toilet and the safety of locomotive operation.

[0046] Therefore, this application uses the entropy weight method and the least squares method to determine the weight of each parameter and the slope of the fitting line in the current analysis period, and takes the result of weighted fusion of the weights of all parameters and the slope of the fitting line as the dynamic risk index of multi-source locomotive toilet related data in the current analysis period.

[0047] It can be understood that fusion can be divided into forward fusion and reverse fusion. This embodiment adopts the forward fusion method. Forward fusion is a fusion method such as addition and multiplication between data. The specific forward fusion method is determined by the implementer according to the actual situation. The application does not impose any special restrictions.

[0048] In this embodiment, the normalized temperature sequence, liquid level sequence, vibration sequence and humidity sequence are used as input, and a dynamic weighted fusion algorithm is used to calculate the weight of each parameter by the entropy weight method: first, the parameter uncertainty is quantified based on the information entropy, and the dynamic allocation of weights of different parameters is realized, and the weights ω1, ω2, ω3, and ω4 of the four parameters are output respectively. The weights adaptively adjust the priority of the contribution of each parameter to the risk of locomotive toilet icing according to the real-time operating environment. For example: if the locomotive starts and stops frequently or passes through bumpy tracks, the vibration causes the liquid level of the locomotive toilet sewage storage tank to fluctuate violently. At this time, the sudden change in the liquid level sequence will trigger the entropy weight method to reduce ω2, avoiding misjudgment of the locomotive toilet emptying need due to short-term liquid level fluctuations, and instead relying on the comprehensive decision-making of temperature and humidity parameters.

[0049] Then, the four data sequences were used as input respectively, and the least squares method was used to output the fitting straight lines of the four data sequences, and the slopes k1, k2, k3, and k4 of the fitting straight lines of the four data sequences were output. The slopes of the fitting straight lines of each parameter reflect the severity of the parameter changes in the locomotive toilet during dynamic operation, which directly affects the real-time performance of the anti-freeze control. Combined with the actual operating status of the locomotive, the specific causes of icing in the locomotive toilet system are analyzed.

[0050] Based on the above analysis, a dynamic risk index A of multi-source locomotive toilet-related data is constructed. The specific calculation formula is:

[0051]

[0052] Specifically, 4 is the parameter type of multi-source locomotive toilet related data, ω i The weight dynamically assigned to the i-th data sequence reflects the impact of the dynamic change of data on the weight allocation through the information entropy quantification parameter uncertainty. The larger its value, the greater the impact of the change of the i-th data on the current icing risk of the locomotive toilet system. Therefore, the most critical locomotive toilet icing inducement (such as high humidity and strong vibration) is automatically focused during locomotive operation to avoid misjudgment caused by static allocation. i is the slope of the fitting line of the i-th data sequence, reflecting the rate of change of the parameter in the historical moment, and its absolute value |k i |Indicates the severity of the change, which is directly related to the actual change of the locomotive toilet parameters. The data slope is converted into the quantification of the icing risk of the locomotive toilet to ensure that the algorithm response corresponds to the physical phenomenon one-to-one. The larger the value, the faster the parameter changes and the easier it is to trigger the risk critical value. The product of the above two parameters is summed, combining the importance of the locomotive toilet parameters and the severity of the change to avoid a single parameter dominating or ignoring weak signals.

[0053] It should be understood that the dynamic risk index A of multi-source locomotive toilet-related data is used to quantify the comprehensive probability of locomotive toilet freezing under the current environment. The larger the A value, the higher the risk of the toilet experiencing a high-risk state under the synergistic effect of multiple parameters.

[0054] Step 3: Combine the long-term risk trends and short-term risk mutations of all dynamic risk indices calculated during all analysis periods, and fuse the two to construct a long-term and short-term dual-modal icing risk index.

[0055] Since the dynamic risk index A solely relies on real-time multiple parameters, it may lead to misjudgment of pipeline freezing risks. For example, short-term fluctuations in the A value may mask long-term hidden dangers, or sudden mutations may not be captured in time. It is difficult to take into account both long-term trends and short-term anomalies, causing problems such as delayed or over-response of anti-freeze control.

[0056] Therefore, the present application combines the long-term risk trends and short-term risk mutations of all dynamic risk indices calculated in all analysis periods collected and integrated to construct a long-term and short-term dual-modal icing risk index.

[0057] In this embodiment, the dynamic risk index A calculated at each historical collection moment is sorted in ascending order of time to obtain a dynamic risk sequence.

[0058] Taking the dynamic risk sequence as input, the sliding average algorithm is first adopted, and the observation time window is set to 5 minutes, corresponding to N = 3000 data points. The 10Hz sampling frequency can balance noise suppression and trend sensitivity, and the step size is 10s to smooth short-term fluctuations. The sliding average value SMA of the dynamic risk sequence is output, which represents the long-term stable trend of the dynamic risk sequence. It can accurately capture the chronic risk accumulation of locomotive toilets in continuous low temperature environments, and highlight the hidden danger of gradual freezing of sewage in locomotive toilets in static retention scenarios.

[0059] The sliding average algorithm is a well-known technique and will not be described in detail. In other embodiments, methods such as exponentially weighted moving average, linear regression, and polynomial regression may also be used to determine the long-term risk trends of all dynamic risk indices.

[0060] Subsequently, the dynamic risk sequence is used as input and the CUSUM cumulative control algorithm is used to output the cumulative deviation value CU through the statistical mean of the dynamic risk index value and the tolerance deviation threshold. This value represents the short-term mutation intensity of the A value and is used to monitor the short-term outbreak risk of locomotive toilets in sudden harsh environments. For example, when a locomotive encounters a blizzard or passes through a high-humidity tunnel, the external humidity sensor data soars, and the dynamic risk index A rises rapidly due to the sudden increase in the humidity parameter weight. The CU algorithm quickly captures this mutation through the cumulative deviation value, indicating an instantaneous surge in the risk of condensation water freezing on the surface of the locomotive toilet.

[0061] The statistical mean is used as a reference value in the CUSUM cumulative control algorithm to measure the difference between the current observed value and the expected value under normal circumstances. The tolerance deviation threshold is set at 1.5 times the standard deviation of all dynamic risk indices and is used in the CUSUM cumulative control algorithm to determine whether the cumulative deviation exceeds the acceptable range, thereby determining whether a significant change or abnormal situation has occurred. The CUSUM cumulative control algorithm is a well-known technology and will not be described in detail. In other embodiments, methods such as moving range, wavelet transform, and difference method can also be used to determine the short-term risk mutation of all dynamic risk indices.

[0062] Based on the above analysis, the long-term and short-term dual-mode icing risk B is constructed. The specific calculation formula is:

[0063] B=Norm(SMA×CU)

[0064] Specifically, SMA is the sliding average of the dynamic risk sequence, which is used to smooth the short-term fluctuations of the dynamic risk index and extract long-term trend characteristics. The locomotive toilet system uses the SMA value to identify the continuous low temperature state, prevent the static sewage in the locomotive toilet system from freezing due to long-term retention, and avoid the hidden freezing problem caused by the traditional system ignoring the long-term trend. It reflects the continuous state of freezing risk. The larger its value, the higher the risk of the toilet in the observation window. The higher the value, the higher the risk of the toilet in the observation window. CU is the cumulative deviation value of the dynamic risk sequence, which is used to detect the positive mutation of the dynamic risk index in the short term and capture sudden risk events. It is used to avoid the sudden freezing of the locomotive toilet in dynamic events such as blizzards and bumps faced by the locomotive. The larger its value, the more the recent dynamic risk index has continuously exceeded the historical normal range and the system is in a rapid deterioration state. Norm () is the normalization function.

[0065] It should be understood that the bimodal icing risk index B is essentially a comprehensive reflection of long-term risk accumulation and short-term risk outbreak, taking into account both "chronic risk" and "acute risk", and dynamically quantifying the complex icing threats faced by locomotive toilets in combination with the locomotive operating status, avoiding the one-sidedness of a single indicator. The larger the value, the higher the icing risk currently faced by the toilet.

[0066] Step 4: Use the dual-modal icing risk index to set a multi-level threshold anti-freeze mechanism strategy.

[0067] Based on the dual-modal icing risk B calculated in step three, a multi-level threshold trigger mechanism is set to achieve graded antifreeze control.

[0068] Specifically, a first threshold B1 and a second threshold B2 are set, where the first threshold is less than the second threshold. The value range of B1 is from 0.2 to 0.4 to avoid frequent activation of heating or draining, with energy conservation being prioritized; the value range of B2 is from 0.6 to 0.8 to balance the emergency response speed and false alarm rate and ensure rapid linkage of full-power anti-freezing measures under high risks. Specifically, in this embodiment, the specific values of B1 and B2 are 0.3 and 0.7.

[0069] When the dual-modal icing risk B < B1, the locomotive toilet is at a low risk level: maintaining the basic monitoring mode, operating the periodic temperature inspection with the lowest energy consumption, continuously updating data through the edge computing unit, and not activating the active heating or draining function;

[0070] When the dual-modal icing risk B1 ≤ B < B2, the locomotive toilet is at a medium risk level: starting the dynamic electric tracing heat strategy, adjusting the heating power according to the slope of the current temperature sequence, and preferentially locally heating the pipeline area with a larger absolute value of the slope to inhibit the icing trend;

[0071] When the dual-modal icing risk B ≥ B2, the locomotive toilet is at a high risk level: immediately activating the emergency draining program, linking the solenoid valve and air pump of the sewage storage tank, optimizing the draining timing in combination with the vibration sensor data (for example, using the mechanical vibration during locomotive operation to assist sewage flow), simultaneously switching to the full-power electric tracing heat mode, and generating an alarm signal on the edge computing platform and pushing it to the on-vehicle monitoring system and remote operation and maintenance center to prompt manual intervention for verification.

[0072] The hierarchical threshold control strategy divides the low, medium, and high risk levels through dual thresholds. On the premise of ensuring the basic monitoring function, it adopts a progressive response mechanism of dynamic electric tracing heat (medium risk) and emergency draining + full-power heating (high risk), reducing energy consumption compared with the traditional continuous full-power heating mode. At the same time, through the design of preferentially heating areas with larger slopes, it effectively alleviates the local temperature difference and microcrack problems caused by the difference in material thermal conductivity. The data preprocessing and real-time analysis capabilities of the edge computing unit ensure the timing optimization of vibration data and draining operations. Using the mechanical vibration during locomotive operation to assist sewage flow not only improves the draining efficiency but also reduces the energy consumption of the air pump, significantly enhancing the average response speed of the anti-freezing system in extreme environments.

[0073] Based on the same inventive concept as the above method, another embodiment of this application also provides an intelligent control system for anti-freezing and draining of locomotive toilets, which includes:

[0074] The Multi-Source Locomotive Toilet Data Collection and Preprocessing Module is used to collect and preprocess multi-source locomotive toilet data from the locomotive toilet system. This module uses a variety of sensors deployed in pipelines, sewage storage tanks, the locomotive chassis, and the exterior of the vehicle to collect real-time environmental and system status data. It also preprocesses the raw data to provide high-quality input for subsequent modules.

[0075] The dynamic risk modeling module uses a dynamic weighted fusion algorithm to dynamically assign weights to multi-source locomotive toilet data and quantify the dynamic risk index of multi-source locomotive toilet data within the current analysis period. This module quantifies the comprehensive probability of the current icing risk based on preprocessed multi-source locomotive toilet data.

[0076] The long-term and short-term risk fusion module is used to combine the long-term risk trends and short-term risk mutations of all dynamic risk indices calculated in all analysis periods collected, and fuse the two to construct a long-term and short-term dual-modal icing risk index.

[0077] The hierarchical antifreeze control module is used to set a multi-level threshold antifreeze mechanism strategy using a dual-modal icing risk index.

[0078] The system adopts a layered modular design and consists of a multi-source locomotive toilet-related data collection and preprocessing module, a dynamic risk modeling module, a long-term and short-term risk fusion module, and a graded anti-freeze control module. Each module is connected in series and works together to achieve a closed-loop process from data collection to risk analysis and intelligent control.

[0079] Each module realizes data interaction and real-time processing through the edge computing unit, forming a closed-loop logic of "data collection → risk modeling → risk fusion → intelligent control", ensuring the system's adaptability and efficient operation in extremely cold environments.

[0080] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not invented herein.

[0081] It will be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A locomotive toilet antifreeze emptying intelligent control method, characterized in that: The method comprises the following steps: Collecting and preprocessing multi-source locomotive toilet-related data in the locomotive toilet system; wherein the multi-source locomotive toilet-related data is collected by multiple sensors deployed in pipelines, sewage storage tanks, locomotive chassis and the outside of the locomotive body; The entropy weight method and the least squares method are used to determine the weight of each type of locomotive toilet-related data and the slope of the fitting line in the current analysis period respectively. The weighted fusion result of all types of locomotive toilet-related data and the absolute value of the slope of the fitting line is used as the dynamic risk index of the multi-source locomotive toilet-related data in the current analysis period. Combining the long-term risk trends and short-term risk mutations of all dynamic risk indices calculated for all analysis periods, and fusing the two, to construct a long-term and short-term dual-modal icing risk index; the long-term risk trend is further determined by extracting the long-term trends of all dynamic risk indices using a sliding average algorithm; A multi-level threshold anti-freeze mechanism strategy is set using the dual-modal icing risk index, including: setting multi-level thresholds to trigger anti-freeze measures at low, medium and high risk levels according to the dual-modal icing risk index; maintaining the basic monitoring mode at the low risk level; starting the dynamic electric heating strategy at the medium risk level; activating the emergency emptying procedure and switching to the full-power electric heating mode at the high risk level.

2. The method for controlling the antifreeze emptying of a locomotive toilet according to claim 1, characterized in that: The multi-source locomotive toilet related data includes: pipeline temperature, sewage level, locomotive vibration and ambient humidity.

3. The method for controlling the antifreeze emptying of a locomotive toilet according to claim 1, characterized in that: The weighted fusion is an operation of performing a weighted summation of the absolute values ​​of the slopes of the fitted straight lines with respect to the weights of all types of data.

4. The method for controlling the antifreeze emptying of a locomotive toilet according to claim 1, characterized in that: The short-term risk mutation is further determined by detecting short-term mutations of all dynamic risk indices using a CUSUM cumulative control algorithm.

5. The method for controlling the antifreeze emptying of a locomotive toilet according to claim 1, characterized in that: The method for constructing the long-term and short-term dual-modal icing risk index is further obtained by multiplying the long-term risk trend and the short-term risk mutation by a normalized result.

6. The method for controlling the antifreeze emptying of a locomotive toilet according to claim 1, characterized in that: The method for setting the multi-level threshold and triggering the anti-freeze measures of low, medium and high risk levels according to the dual-modal icing risk index is as follows: Setting a first threshold and a second threshold; wherein the first threshold is less than the second threshold; When the bimodal icing risk is less than the first threshold, the locomotive toilet is at a low risk level; When the bimodal icing risk is greater than or equal to the first threshold and less than the second threshold, the locomotive toilet is at a medium risk level; When the bimodal icing risk is greater than or equal to the second threshold, the locomotive toilet is at a high risk level.

7. An intelligent control system for antifreeze emptying of locomotive toilets, characterized in that: To implement the intelligent control method for antifreeze emptying of a locomotive toilet according to any one of claims 1 to 6, the system comprises: Multi-source locomotive toilet related data collection and preprocessing module, used to collect multi-source locomotive toilet related data in the locomotive toilet system and perform preprocessing; A dynamic risk modeling module is used to dynamically assign weights to multi-source locomotive toilet-related data using a dynamic weighted fusion algorithm, and quantify the dynamic risk index of multi-source locomotive toilet-related data within the current analysis period; The long-term and short-term risk fusion module is used to combine the long-term risk trend and short-term risk mutation of all dynamic risk indices calculated in all analysis periods, and fuse the two to construct a long-term and short-term dual-modal icing risk index; The hierarchical antifreeze control module is used to set a multi-level threshold antifreeze mechanism strategy using a dual-modal icing risk index.

Images