Full-automatic centralized cutting fluid supply control method and system

Through real-time monitoring and closed-loop control system of sensor arrays, combined with multi-stage filtration, centrifugal separation and machine learning models, the problems of unstable supply pressure and difficult to accurately regulate the concentration in traditional cutting fluid management are solved, and efficient circulation purification of cutting fluid and intelligent management of equipment are achieved, and production efficiency and equipment life are improved.

CN120507958AActive Publication Date: 2025-08-19ZHUZHOU SIXING MACHINERY

Patent Information

Application Number
CN202510611611.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-05-13
Publication Date
2025-08-19
Estimated Expiration
2045-05-13

AI Technical Summary

Technical Problem

Traditional cutting fluid management methods rely on manual operations, making it difficult to achieve stable fluid supply pressure, precise concentration regulation and efficient purification, resulting in high resource waste and operation and maintenance costs, making it difficult to adapt to the needs of large-scale and complex production scenarios.

Method used

The cutting fluid concentration and liquid supply system pressure are monitored in real time through the sensor array, real-time data is generated using high-frequency sampling technology, and a closed-loop control system is formed using proportional-integration-differential algorithm, combining multi-stage filtration and centrifugal separation devices to purify the cutting fluid, and using machine learning models to analyze the wear trend of the equipment, and dynamically adjust the liquid supply parameters.

Benefits of technology

It realizes stable pressure of cutting fluid supply, precise concentration regulation and efficient circulation purification, extends the service life of the equipment, improves processing quality and production efficiency, and reduces maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120507958A_ABST
    Figure CN120507958A_ABST
Patent Text Reader

Abstract

The invention discloses a cutting fluid full-automatic centralized liquid supply control method and system, cutting fluid concentration data and liquid supply system pressure data are obtained through a sensor array, real-time monitoring data are generated by adopting a high-frequency sampling technology, and if the cutting fluid concentration data deviates from a preset threshold range or the liquid supply system pressure data exceeds a stable range, the cutting fluid is monitored. If so, generating an abnormal signal; according to the abnormal signal, the operation frequency of a liquid supply pump is adjusted through a proportional-integral-differential algorithm, a closed-loop control system is formed to maintain a target pressure value, and meanwhile the opening value of a liquid distribution valve is adjusted according to the concentration deviation signal; and the cutting fluid return liquid is extracted from the adjusted liquid supply system, the cutting fluid return liquid is treated through a multi-stage filtering device and a centrifugal separation device, and if it is detected that the impurity content of the cutting fluid return liquid is lower than a preset threshold value, it is judged that purification treatment is completed. The stable performance of the cutting fluid can be kept, the service life of equipment is prolonged, the processing quality and the production efficiency are improved, and the cutting fluid has important practical value.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular discloses a fully automatic centralized cutting fluid supply control method and system. Background Art

[0002] Cutting fluid management is a crucial component of modern manufacturing, directly impacting machining accuracy, equipment life, and production efficiency. In the context of intelligent manufacturing and green production, automated cutting fluid supply and circulation purification have become key technologies for enhancing factory competitiveness. However, traditional cutting fluid management methods have significant limitations. They rely on manual operation, lack outdated monitoring methods, and have low purification efficiency, resulting in wasted resources and high operation and maintenance costs. Frequent manual intervention, in particular, makes it difficult to adapt to the needs of large-scale, complex production scenarios. Furthermore, the lack of real-time data support makes it difficult to achieve precise control and fault prediction.

[0003] The core challenge lies in automating the entire process of cutting fluid supply, monitoring, and purification. Cutting fluid supply requires stable pressure output to ensure machining quality, but traditional systems experience large pressure fluctuations, making it difficult to meet the demands of high-precision machining. Unstable pressure control further hinders precise maintenance of the fluid concentration. Concentration deviations can affect the lubrication and cooling properties of the cutting fluid, increasing equipment wear. Inadequate concentration management directly impacts return fluid purification. Traditional filtration technologies struggle to effectively remove fine impurities and contaminated oil, shortening the life of the cutting fluid. These intertwined factors constitute the core technical challenge of automated management.

[0004] Therefore, how to achieve stable cutting fluid supply pressure, precise concentration control and efficient circulation purification through the integration of intelligent sensing, closed-loop control and multi-stage purification technology has become a key issue in promoting the upgrading of intelligent manufacturing. Summary of the Invention

[0005] The present invention provides a fully automatic centralized cutting fluid supply control method and system, aiming to solve at least one defect existing in the above-mentioned prior art.

[0006] One aspect of the present invention relates to a fully automatic centralized cutting fluid supply control method, comprising the following steps:

[0007] The sensor array acquires cutting fluid concentration data and fluid supply system pressure data, and uses high-frequency sampling technology to generate real-time monitoring data. If the cutting fluid concentration data deviates from the preset threshold range or the fluid supply system pressure data exceeds the stable range, an abnormal signal is generated;

[0008] According to the abnormal signal, the proportional-integral-differential algorithm is used to adjust the operating frequency of the liquid supply pump to form a closed-loop control system to maintain the target pressure value. At the same time, the opening value of the liquid distribution valve is adjusted according to the concentration deviation signal;

[0009] Extracting cutting fluid return from the regulated fluid supply system, processing the cutting fluid return through a multi-stage filtration device and a centrifugal separation device. If the impurity content of the cutting fluid return is detected to be lower than a preset threshold, the purification process is determined to be complete.

[0010] A machine learning model is used to analyze the characteristics of the purified cutting fluid performance parameters and equipment operation data, and a pre-established fault prediction model is used to generate equipment wear trend prediction results;

[0011] Adjust the cutting fluid circulation purification frequency and fluid supply parameters based on the equipment wear trend prediction results. If the equipment wear is predicted to intensify, update the purification frequency parameters and optimize the fluid supply pressure setting value.

[0012] Furthermore, the cutting fluid concentration data and the fluid supply system pressure data are acquired through the sensor array, and high-frequency sampling technology is used to generate real-time monitoring data. If the cutting fluid concentration data deviates from a preset threshold range or the fluid supply system pressure data exceeds a stable range, the step of generating an abnormal signal includes:

[0013] By using high-frequency sampling technology through the sensor array, cutting fluid concentration data and fluid supply system pressure data are obtained from the processing equipment, generating real-time monitoring data streams, which are stored in a pre-established database to obtain continuous time series data;

[0014] According to the continuous time series data, the cutting fluid concentration data is compared with the preset threshold range. If the cutting fluid concentration data exceeds the preset threshold range, it is marked as a concentration abnormal state and a concentration abnormality mark is obtained;

[0015] According to the continuous time series data, the pressure data of the liquid supply system is compared with the stable range. If the pressure data of the liquid supply system exceeds the stable range, it is marked as a pressure abnormal state and a pressure abnormality flag is obtained;

[0016] The concentration abnormality flag and the pressure abnormality flag are obtained through the signal generation logic. If at least one of the concentration abnormality flag and the pressure abnormality flag is in an abnormal state, an abnormal signal is generated.

[0017] Furthermore, the steps of adjusting the operating frequency of the liquid supply pump using a proportional-integral-differential algorithm based on the abnormal signal to form a closed-loop control system to maintain the target pressure value, and adjusting the opening value of the liquid distribution valve based on the concentration deviation signal include:

[0018] According to the abnormal signal, the control algorithm calculates the operating frequency adjustment amount of the liquid supply pump and the opening adjustment amount of the liquid distribution valve, and generates frequency control instructions and opening control instructions;

[0019] The frequency control instruction is used to adjust the operating frequency of the liquid supply pump, and the opening value of the liquid distribution valve is adjusted through the opening control instruction. A closed-loop control data stream is generated and stored in the database to obtain real-time adjustment records.

[0020] Further, the steps of extracting cutting fluid return from the adjusted fluid supply system, treating the cutting fluid return through a multi-stage filtration device and a centrifugal separation device, and determining that the purification process is completed if the impurity content of the cutting fluid return is detected to be lower than a preset threshold value include:

[0021] Extract cutting fluid return from the fluid supply system through a return fluid collection device, use a flow sensor to obtain flow data of the cutting fluid return, and obtain a return fluid collection data stream;

[0022] The cutting fluid return corresponding to the return fluid collection data stream is processed through a multi-stage filtration device. The first-stage filter is used to remove large particles of impurities, and the second-stage filter is used to remove small particles to obtain the primary filtrate.

[0023] The primary filtrate is treated by a centrifugal separation device, and the residual particles and liquid are separated by a centrifuge at a constant speed to obtain a secondary purified liquid;

[0024] The impurity content data of the secondary purification liquid is obtained by detecting the sensor. If the impurity content is lower than the preset threshold, the purification process is judged to be completed and a purification completion data stream is obtained.

[0025] Furthermore, a machine learning model is used to perform feature analysis on the performance parameters of the purified cutting fluid and the equipment operation data. The steps of generating equipment wear trend prediction results using a pre-established fault prediction model include:

[0026] The cutting fluid performance parameters and equipment operation data are acquired from the equipment operation environment through the sensor data acquisition device. The flow sensor is used to collect the cutting fluid flow value, the temperature sensor is used to collect the cutting fluid temperature value, and the vibration sensor is used to collect the equipment vibration frequency data to obtain the original collected data stream;

[0027] The data preprocessing module is used to process the original collected data stream. The cutting fluid flow value, cutting fluid temperature value and equipment vibration frequency data are denoised by the mean filtering method. The denoised data is then dimensionally unified by the standardization method to obtain the preprocessed data stream.

[0028] The pre-processed data stream is analyzed using a random forest algorithm to extract the fluidity characteristics of the cutting fluid flow value, the thermal stability characteristics of the cutting fluid temperature value, and the equipment wear characteristics of the equipment vibration frequency data, thereby obtaining a feature vector set.

[0029] The feature vector set is processed through a pre-established fault prediction model, and the fluidity characteristics, thermal stability characteristics and equipment wear characteristics are classified and predicted using the logistic regression method. If the prediction probability is higher than the preset threshold, it is judged that the equipment has a wear trend, and the wear trend prediction data stream is obtained.

[0030] Furthermore, the circulation purification frequency and fluid supply parameters of the cutting fluid are adjusted according to the equipment wear trend prediction result. If the equipment wear is predicted to intensify, the purification frequency parameters are updated and the fluid supply pressure setting value is optimized. The steps include:

[0031] Adjust cutting fluid cycle parameters based on wear trend prediction data stream;

[0032] Determine whether the equipment wear is aggravated. If it is predicted that the equipment wear is aggravated, the linear regression method is used to optimize the liquid supply pressure setting value to obtain the adjusted purification frequency parameters and liquid supply pressure value.

[0033] Another aspect of the present invention relates to a fully automatic centralized cutting fluid supply control system for implementing the above-mentioned fully automatic centralized cutting fluid supply control method. The fully automatic centralized cutting fluid supply control system comprises:

[0034] The first generation module is used to obtain cutting fluid concentration data and fluid supply system pressure data through a sensor array, and use high-frequency sampling technology to generate real-time monitoring data. If the cutting fluid concentration data deviates from a preset threshold range or the fluid supply system pressure data exceeds a stable range, an abnormal signal is generated;

[0035] The adjustment module is used to adjust the operating frequency of the liquid supply pump based on the abnormal signal using the proportional-integral-differential algorithm to form a closed-loop control system to maintain the target pressure value, and at the same time adjust the opening value of the liquid distribution valve based on the concentration deviation signal;

[0036] a determination module, configured to extract cutting fluid return from the regulated fluid supply system, process the cutting fluid return through a multi-stage filtration device and a centrifugal separation device, and determine that purification processing is complete if the impurity content of the cutting fluid return is detected to be lower than a preset threshold;

[0037] The second generation module is used to use a machine learning model to perform feature analysis on the performance parameters of the purified cutting fluid and the equipment operation data, and to generate equipment wear trend prediction results through a pre-established fault prediction model;

[0038] The update module is used to adjust the circulation purification frequency and fluid supply parameters of the cutting fluid according to the equipment wear trend prediction results. If the equipment wear is predicted to increase, the purification frequency parameters are updated and the fluid supply pressure setting value is optimized.

[0039] Furthermore, the first generation module includes:

[0040] The first acquisition unit is used to acquire cutting fluid concentration data and fluid supply system pressure data from the processing equipment through a sensor array using high-frequency sampling technology, generate a real-time monitoring data stream, and store it in a pre-established database to obtain continuous time series data;

[0041] The second acquisition unit is configured to compare the cutting fluid concentration data with a preset threshold range based on the continuous time series data, and if the cutting fluid concentration data exceeds the preset threshold range, mark it as a concentration abnormality state and obtain a concentration abnormality flag;

[0042] a third acquiring unit, configured to compare the pressure data of the liquid supply system with a stable range based on the continuous time series data, and if the pressure data of the liquid supply system exceeds the stable range, mark it as a pressure abnormality state and obtain a pressure abnormality flag;

[0043] The first generating unit is configured to obtain a concentration abnormality flag and a pressure abnormality flag through signal generating logic, and generate an abnormal signal if at least one of the concentration abnormality flag and the pressure abnormality flag is in an abnormal state.

[0044] Furthermore, the adjustment module includes:

[0045] The second generating unit is used to calculate the operating frequency adjustment amount of the liquid supply pump and the opening adjustment amount of the liquid distribution valve through a control algorithm according to the abnormal signal, and generate a frequency control instruction and an opening control instruction;

[0046] The fourth acquisition unit is used to adjust the operating frequency of the liquid supply pump using the frequency control instruction, adjust the opening value of the liquid distribution valve through the opening control instruction, generate a closed-loop control data stream, store it in the database, and obtain real-time adjustment records.

[0047] Furthermore, the determination module includes:

[0048] A fifth acquisition unit is used to extract cutting fluid return from the fluid supply system through the return fluid collection device, and use a flow sensor to obtain flow data of the cutting fluid return to obtain a return fluid collection data stream;

[0049] A sixth acquisition unit is used to process the cutting fluid return corresponding to the return fluid collection data stream through a multi-stage filtration device, using a first-stage filter to remove large particles of impurities, and then using a second-stage filter to remove small particles to obtain a primary filtrate;

[0050] a seventh obtaining unit for processing the primary filtrate by a centrifugal separation device, using a centrifuge at a constant speed to separate residual particles from the liquid to obtain a secondary purified liquid;

[0051] The eighth acquisition unit is used to acquire the impurity content data of the secondary purification liquid through the detection sensor. If the impurity content is lower than a preset threshold, it is determined that the purification process is completed, and a purification completion data stream is obtained.

[0052] The beneficial effects achieved by the present invention are:

[0053] The present invention provides a fully automatic centralized cutting fluid supply control method and system. A sensor array monitors the cutting fluid concentration and supply pressure in real time. When an anomaly is detected, the frequency of the supply pump and the opening of the dispensing valve are automatically adjusted to form a closed-loop control system. Simultaneously, the recovered cutting fluid undergoes multi-stage filtration and centrifugal purification, and a machine learning model is used to analyze cutting fluid performance parameters and equipment operating data to predict equipment wear trends. Based on the prediction results, the cutting fluid circulation purification frequency and supply parameters are dynamically adjusted to achieve intelligent management of the cutting fluid and preventive maintenance of equipment wear. The present invention can maintain stable cutting fluid performance, extend the service life of equipment, and improve processing quality and production efficiency, thus possessing significant practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 The figure is a flow chart of an embodiment of a fully automatic centralized cutting fluid supply control method of the present invention. DETAILED DESCRIPTION

[0055] In order to better understand the above technical solution, the above technical solution will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0056] like Figure 1 As shown, the first embodiment of the present invention provides a fully automatic centralized cutting fluid supply control method, comprising the following steps:

[0057] Step S100: obtain cutting fluid concentration data and fluid supply system pressure data through the sensor array, and use high-frequency sampling technology to generate real-time monitoring data. If the cutting fluid concentration data deviates from the preset threshold range or the fluid supply system pressure data exceeds the stable range, an abnormal signal is generated.

[0058] Cutting fluid concentration data refers to the volume or mass ratio of the effective ingredients (such as emulsified oil, additives, etc.) and diluent (usually water) in the cutting fluid. It is used to quantify the mixing ratio of the cutting fluid and directly affects the cooling, lubrication, rust prevention and other properties.

[0059] The pressure data of the fluid supply system refers to the continuous pressure value per unit area generated by the fluid in the pipeline when the cutting fluid is transported in the centralized fluid supply system. It is usually measured in megapascals (MPa) and is used to quantify the system's liquid delivery capacity and stability to the processing equipment.

[0060] High-frequency sampling technology refers to the digital process of continuously and rapidly collecting physical quantities at a rate much higher than the target signal change frequency (usually ≥1kHz). By acquiring dense data points in a short period of time, it can accurately capture the instantaneous change characteristics of dynamic systems. It is a key technical means in the fields of industrial automation and precision monitoring.

[0061] Real-time monitoring data refers to the continuous collection, processing and feedback of key system parameters (such as pressure, concentration, temperature, etc.) at a response speed of milliseconds to seconds through sensors, instruments and other hardware equipment during industrial production or equipment operation, forming a dynamic closed-loop control chain to ensure that process parameters are within the preset threshold range in real time.

[0062] Abnormal signals refer to parameter fluctuations, mutations or abnormal characteristics that deviate from preset thresholds or normal states and are detected by sensors or monitoring devices during the operation of industrial systems (such as cutting fluid supply systems and hydraulic systems). They usually manifest as pressure abnormalities, concentration deviations, contaminant mixing, equipment component failure, etc., and require triggering early warnings or control interventions to ensure system stability and safety.

[0063] Step S200: According to the abnormal signal, the proportional-integral-differential algorithm is used to adjust the operating frequency of the liquid supply pump to form a closed-loop control system to maintain the target pressure value, and at the same time, the opening value of the liquid distribution valve is adjusted according to the concentration deviation signal.

[0064] The Proportional-Integral-Derivative (PID) algorithm is a closed-loop control algorithm based on real-time feedback of an error signal. By calculating a linear combination of the proportional term (P), the integral term (I), and the differential term (D), it dynamically adjusts the system output to eliminate deviations between the actual value and the set target. The core goal of the PID algorithm is to achieve precise and stable control of controlled parameters (such as temperature, pressure, and flow) through a negative feedback mechanism.

[0065] A closed-loop control system (CLCS) is an automatic adjustment system based on a real-time feedback mechanism. This system continuously collects the output signal of the controlled object, dynamically compares it with a preset target value (setpoint), calculates the error, and uses a control algorithm (such as PID) to generate correction instructions. This control architecture drives the actuator to adjust the input, ultimately stabilizing the output within the target range. The core feature of a closed-loop control system is the closed-loop logic of "detection → feedback → correction." It possesses interference resistance and self-adaptation capabilities and is widely used in industrial automation, robotics, process control, and other fields.

[0066] The target pressure value is the expected pressure parameter preset in the industrial control system. It serves as the input set value of the closed-loop control process (such as the set pressure of 0.3MPa in a constant-pressure water supply system). The actual pressure is fed back in real time by the sensor and compared with the target value. The actuator (such as a variable frequency pump or valve) is driven to dynamically adjust the output, ultimately stabilizing the system pressure within the target range.

[0067] The concentration deviation signal is a key parameter in a closed-loop control system that characterizes the difference between the actual concentration and the target concentration. It is defined as the difference between the set concentration value and the actual concentration value detected by the sensor in real time.

[0068] The opening value of the liquid dispensing valve is a quantitative parameter that characterizes the size of the valve's flow cross-sectional area. It is defined as the percentage of the valve's displacement or rotation angle from fully closed to fully open to the maximum stroke. It is used to accurately control the flow, pressure and mixing ratio of fluids (such as liquid medicines and solvents).

[0069] Step S300: extracting cutting fluid return from the adjusted fluid supply system, and processing the cutting fluid return through a multi-stage filtration device and a centrifugal separation device. If it is detected that the impurity content of the cutting fluid return is lower than a preset threshold, it is determined that the purification process is completed.

[0070] Cutting fluid return refers to a closed-loop management system that recovers, filters, purifies and reuses used cutting fluid during the metalworking process.

[0071] A multi-stage filtration system uses a series of filter units with varying precision and functionality to progressively remove impurities (such as metal debris, oil, and microorganisms) from cutting fluids or industrial fluids, achieving efficient purification and recycling. The core design of a multi-stage filtration system balances filtration efficiency and operating costs through a progressive filtration mechanism, ensuring that the fluid meets the cleanliness requirements of the processing process throughout its circulation.

[0072] A centrifugal separator is a mechanical device that uses centrifugal force to efficiently separate mixed fluids (such as liquid-solid, liquid-liquid or liquid-gas). Its core principle is to use the inertial centrifugal force generated by high-speed rotation to cause components of different densities or particle sizes to move differently, thereby achieving stratification, sedimentation or filtration.

[0073] Purification refers to a systematic process that removes contaminants (such as suspended particles, oil, microorganisms, or toxic substances) from fluids (such as water, cutting fluids, and industrial wastewater) through physical, chemical, or biological methods to ensure that the fluid meets specific usage standards or environmental emission requirements.48 The core goal of purification is to separate and retain impurities, restore the fluid's functional properties, or reduce its environmental hazards.

[0074] Step S400: Use a machine learning model to perform feature analysis on the performance parameters of the purified cutting fluid and the equipment operation data, and generate equipment wear trend prediction results through a pre-established fault prediction model.

[0075] A machine learning model is a mathematical function or structure built based on data and algorithms. Its core function is to learn the patterns and features in a training dataset and establish a mapping between input variables and output results, thereby enabling prediction, classification, or decision-making for unknown data. The essence of a machine learning model is to optimize prediction errors by adjusting internal parameters (such as weights and biases), ultimately forming a generalizable rule system.

[0076] Cutting fluid performance parameters are a set of indicators used to quantify the functional performance of cutting fluids during the machining process. They cover core characteristics such as lubricity, cooling, stability, and rust resistance, and directly affect machining efficiency, tool life, and workpiece surface quality.

[0077] Equipment operation data refers to a collection of quantitative information captured through sensors, control systems, or manual recordings, reflecting the real-time status and performance parameters of mechanical equipment during operation. This information covers key indicators such as equipment energy consumption, workload, environmental parameters, and operating efficiency. The core function of equipment operation data is to enable equipment health monitoring, fault warnings, and process optimization through data collection and analysis.

[0078] A fault prediction model is a quantitative tool built using machine learning, statistical analysis, or physical modeling techniques based on real-time or historical equipment operating data (such as sensor parameters like temperature, vibration, and current). Its core goal is to analyze equipment performance degradation patterns and failure characteristics to predict future failure probabilities and remaining useful life (RUL), providing decision support for preventive maintenance.

[0079] Equipment wear trends refer to the regular degradation of a device's physical properties over time, whether in use or idle. This is manifested by significant variations in wear rate at different stages. This trend is typically quantified using a wear-over-time curve, reflecting the dynamic evolution of equipment from initial use to functional failure. It serves as a core basis for developing preventive maintenance strategies.

[0080] Step S500: Adjust the circulation purification frequency and fluid supply parameters of the cutting fluid according to the equipment wear trend prediction result. If it is predicted that the equipment wear will increase, update the purification frequency parameter and optimize the fluid supply pressure setting value.

[0081] Purification cycle frequency refers to the number of complete processes a specific system or device completes per unit of time to filter pollutants, renew water quality, or improve air quality. Its core goal is to maintain the cleanliness, safety, and functionality of environmental media through periodic purification, with specific manifestations varying depending on the application scenario.

[0082] Fluid supply parameters refer to the key control indicators set during industrial processing to ensure the effective delivery of cutting fluid or grinding fluid and maintain its performance. They cover the physical and chemical properties of the liquid, the operating parameters of the circulation system and the requirements for pollutant control, and directly affect the processing quality, equipment life and cost-effectiveness.

[0083] Fluid supply parameters refer to the key control indicators set during industrial processing to ensure the effective delivery of cutting fluid or grinding fluid and maintain its performance. They cover the physical and chemical properties of the liquid, the operating parameters of the circulation system and the requirements for pollutant control, and directly affect the processing quality, equipment life and cost-effectiveness.

[0084] The fluid supply pressure setpoint is the pre-set target pressure in industrial fluid supply systems (such as cutting fluid and coolant circulation systems) to ensure effective fluid coverage of the machining area, stable chip removal, and optimal performance. This pressure setpoint directly impacts fluid delivery efficiency, surface finish quality, and energy consumption, and must be dynamically optimized based on pipe network resistance, flow requirements, and process characteristics.

[0085] Furthermore, the fully automatic centralized cutting fluid supply control method provided in this embodiment includes step S100:

[0086] Step S110: Using high-frequency sampling technology through a sensor array, obtain cutting fluid concentration data and fluid supply system pressure data from the processing equipment, generate real-time monitoring data streams, and store them in a pre-established database to obtain continuous time series data.

[0087] In one possible implementation, a sensor array uses high-frequency sampling technology to acquire cutting fluid concentration and fluid supply system pressure data from the processing equipment, forming a real-time monitoring data stream. For example, the sensor collects concentration and pressure data at a frequency of 100 times per second, ensuring high temporal resolution of the data stream.

[0088] The cutting fluid concentration is measured by an optical refraction sensor, based on the linear relationship between refractive index and concentration, with a typical concentration range of 5% to 15%.

[0089] The pressure of the liquid supply system is obtained through a piezoelectric sensor, and the normal pressure range is 2 to 5 bar.

[0090] The collected data is transmitted to the database via the Industrial Internet of Things protocol, generating a continuous time series. This high-frequency sampling can capture transient changes and significantly improve the timeliness of anomaly detection.

[0091] Specifically, when storing time series data in a database, a time series database is used to optimize storage efficiency. For example, each record contains a timestamp, concentration value, and pressure value, such as 2025-05-03 10:00:00, concentration 7.8%, pressure 3.2 bar. The database supports queries within seconds, facilitating subsequent analysis. The advantage of storing continuous time series is that historical trends can be traced, providing a data foundation for analyzing the causes of anomalies.

[0092] Step S120: Compare the cutting fluid concentration data with a preset threshold range based on the continuous time series data. If the cutting fluid concentration data exceeds the preset threshold range, it is marked as an abnormal concentration state, and an abnormal concentration flag is obtained.

[0093] In one embodiment, the cutting fluid concentration is compared to a preset threshold range of 5% to 15%. If the concentration is below 5% or above 15%, it is flagged as an abnormal concentration state. For example, if the concentration of a sample is 4.2%, which is below the threshold, the system generates an abnormal concentration flag. Low concentrations may lead to insufficient lubrication and increased tool wear, while high concentrations may cause foaming and affect machining accuracy. Generating an abnormal concentration flag helps quickly locate problems and reduce the risk of equipment damage.

[0094] Step S130: Compare the pressure data of the liquid supply system with the stable range based on the continuous time series data. If the pressure data of the liquid supply system exceeds the stable range, it is marked as a pressure abnormality state, and a pressure abnormality flag is obtained.

[0095] Preferably, the pressure data from the liquid supply system is compared to a stable range of 2 to 5 bar. If the pressure exceeds this range, such as a detected pressure of 5.8 bar, a pressure anomaly is flagged. Excessively high pressure may be caused by a clogged pipeline, while low pressure may be due to a pump failure. The generation of a pressure anomaly indicator provides a timely warning and prevents process interruptions.

[0096] It should be noted that pressure anomaly detection must be combined with historical data trends to avoid false alarms due to short-term fluctuations. For example, the flag will only be triggered if the pressure exceeds the range for 5 consecutive seconds.

[0097] Step S140: Obtain the concentration abnormality flag and the pressure abnormality flag through signal generation logic, and generate an abnormal signal if at least one of the concentration abnormality flag and the pressure abnormality flag is in an abnormal state.

[0098] For example, the signal generation logic receives a concentration abnormality flag and a pressure abnormality flag, and determines whether to generate an abnormality signal through a logical OR operation. In one embodiment, the concentration is 4.2% and the pressure is 3.5 bar. Only the concentration is abnormal, and the system generates an abnormality signal. In another example, the concentration is 7.8% and the pressure is 5.8 bar. Only the pressure is abnormal, and an abnormality signal is still generated. If both are abnormal, such as a concentration of 4.2% and a pressure of 5.8 bar, a signal is also generated. This logic ensures that any single abnormality can trigger a response, improving the robustness of the system. The abnormality signal can be sent to the operator interface through an audible and visual alarm or prompted to adjust the concentration or check the liquid supply system in a timely manner.

[0099] It is understandable that the advantage of the above method is that it can achieve accurate monitoring of cutting fluid concentration and supply pressure through high-frequency sampling, real-time comparison and logical judgment.

[0100] For example, one machining workshop avoided a 30% reduction in tool life by promptly detecting concentration anomalies and adjusting the cutting fluid ratio. Another example is the use of pressure anomaly warnings, which enabled early detection of pump failures and reduced downtime by two hours. These benefits have significantly improved machining efficiency and equipment reliability, while reducing maintenance costs.

[0101] Furthermore, the fully automatic centralized cutting fluid supply control method provided in this embodiment includes step S200:

[0102] Step S210: Calculate the operating frequency adjustment amount of the liquid supply pump and the opening adjustment amount of the liquid distribution valve through a control algorithm according to the abnormal signal, and generate a frequency control instruction and an opening control instruction.

[0103] Preferably, the abnormal signal triggers a control algorithm to calculate the adjustment amount for the operating frequency of the liquid supply pump and the opening adjustment amount of the liquid distribution valve. For example, when the concentration is low, the control algorithm increases the opening of the liquid distribution valve to replenish the concentrate; when the pressure is high, the control algorithm reduces the frequency of the liquid supply pump to reduce the pipeline load.

[0104] In one example, at a concentration of 5.8%, the algorithm generates a command to increase the opening by 10%, while at a pressure of 4.8 bar, it generates a command to reduce the frequency by 5Hz. These commands are transmitted to the actuator via industrial Ethernet, ensuring precise regulation.

[0105] Step S220: Use the frequency control instruction to adjust the operating frequency of the liquid supply pump, adjust the opening value of the liquid distribution valve through the opening control instruction, generate a closed-loop control data stream, store it in the database, and obtain real-time adjustment records.

[0106] It's easy to understand that frequency control commands and aperture control commands drive the liquid supply pump and liquid distribution valve, forming a closed-loop control data flow. For example, if the liquid supply pump frequency is adjusted from 60Hz to 55Hz, the liquid distribution valve aperture will increase from 50% to 60%. The relevant data will be stored in the database, with a record such as 2025-05-03 10:00:01, frequency 55Hz, aperture 60%. This closed-loop control stabilizes system operation through real-time feedback.

[0107] It's important to note that the closed-loop control data stream is stored in a database, creating a real-time adjustment record that supports subsequent traceability. For example, a concentration adjustment from 5.8% to 6.5% and a pressure recovery from 4.8 bar to 3.5 bar were recorded as taking 10 seconds. This record facilitates analysis of system response speed and stability, improving maintenance efficiency.

[0108] In one embodiment, the control algorithm optimizes the adjustment amount by integrating historical data. For example, it analyzes concentration trends over the past hour to predict changes in the valve opening and avoid overshoot. This data-driven approach enhances control accuracy and reduces manual intervention.

[0109] Furthermore, the fully automatic centralized cutting fluid supply control method provided in this embodiment includes step S300:

[0110] Step S310: extract cutting fluid return from the fluid supply system through a return fluid collection device, use a flow sensor to obtain flow data of the cutting fluid return, and obtain a return fluid collection data stream.

[0111] For example, a return fluid collection device is used in a fluid supply system to extract cutting fluid return. Its core function is to provide data support for subsequent processing through accurate flow monitoring. A flow sensor collects return fluid flow data in real time, forming a return fluid collection data stream.

[0112] For example, a factory's liquid supply system returns 200 liters of liquid per minute. A flow sensor records data at a one-second interval, generating a data stream such as "2025-05-03 10:00:01 flow rate 198 liters / minute." This data stream provides the basis for subsequent filtration and purification, ensuring that the system can adjust its treatment strategy based on actual return flow conditions.

[0113] Step S320: Process the cutting fluid return corresponding to the return fluid collection data stream through a multi-stage filtration device, use a first-stage filter to remove large particles of impurities, and then use a second-stage filter to remove small particles to obtain a primary filtrate.

[0114] In one possible implementation, a multi-stage filtration device performs graded treatment on the cutting fluid return. The first-stage filter usually uses a metal mesh with a larger pore size, which is used for large particle impurities such as metal chips. Preferably, in a certain embodiment, the pore size of the first-stage filter is 0.5 mm, which can effectively intercept particles with a diameter greater than 0.5 mm. The second-stage filter uses a finer filter material with a pore size reduced to 0.05 mm, which is used for tiny particles such as grinding chips. Specifically, the turbidity of the primary filtrate is reduced from the initial 500NTU to 100NTU, significantly improving the cleanliness of the liquid. This graded filtration method ensures efficient impurity removal through a step-by-step pore size design.

[0115] Step S330: Process the primary filtrate through a centrifugal separation device, using a centrifuge at a constant speed to separate the residual particles from the liquid to obtain a secondary purified liquid.

[0116] It will be appreciated that the centrifugal separation device further processes the primary filtrate by separating the remaining particles through high-speed rotation.

[0117] The centrifuge runs at a constant speed, for example 3000 rpm, and uses centrifugal force to throw the denser particles to the outside, separating out the clear secondary purified liquid.

[0118] In one embodiment, the impurity content of the secondary purified liquid after centrifugation is reduced from 100 mg / L to 20 mg / L. This constant speed design avoids uneven separation due to speed fluctuations and improves purification stability.

[0119] Step S340: Obtain impurity content data of the secondary purification liquid through a detection sensor. If the impurity content is lower than a preset threshold, it is determined that the purification process is completed, and a purification completion data stream is obtained.

[0120] It's important to note that the detection sensor monitors the impurity content of the secondary purification fluid in real time to determine whether it meets the purification standard. For example, if the preset impurity content threshold is 10 mg / L and the detection value is 8 mg / L, a purification completion data stream is generated, recording something like "Impurity content 8 mg / L at 2025-05-03 10:00:05." Conversely, if the detection value is 15 mg / L, the system triggers additional processing cycles until the standard is met.

[0121] In one embodiment, the detection sensor utilizes laser turbidity analysis technology, accurately measuring impurity levels through the principle of light scattering, ensuring reliable results. Specifically, the generation of return liquid collection data streams and purification completion data streams provides traceability for system operation. For example, during one treatment, the flow sensor recorded a return liquid flow rate of 195 liters / minute, and the detection sensor confirmed that the impurity content had dropped to 9 mg / L. The entire treatment process took 30 seconds, and the data stream was stored in a database. This data record facilitates subsequent analysis of system efficiency and optimization of the processing flow.

[0122] Optimally, a combination of multi-stage filtration and centrifugal separation, coupled with sensor monitoring, forms a complete purification chain. For example, one factory used this system to increase the recycling rate of cutting fluid from 60% to 85%, significantly reducing fluid waste. This chain design, through multi-level processing and precise detection, ensures efficient purification of cutting fluid and safeguards the stable operation of the fluid supply system.

[0123] Furthermore, the fully automatic centralized cutting fluid supply control method provided in this embodiment includes step S400:

[0124] Step S410: obtain cutting fluid performance parameters and equipment operation data from the equipment operation environment through the sensor data acquisition device, use the flow sensor to collect the cutting fluid flow value, use the temperature sensor to collect the cutting fluid temperature value, and use the vibration sensor to collect the equipment vibration frequency data to obtain the original collected data stream.

[0125] For example, in cutting fluid management, the core of the sensor data acquisition device is to obtain key performance parameters in real time.

[0126] Flow sensors monitor the amount of cutting fluid circulating, temperature sensors record temperature changes, and vibration sensors capture frequency signals during equipment operation. These sensors work together to form a multidimensional data stream. For example, in a factory's processing equipment, a flow sensor records the cutting fluid flow rate every second, generating data such as "2025-05-03 10:00:01: Flow rate 180 liters / minute"; a temperature sensor simultaneously records a temperature of 45 degrees Celsius; and a vibration sensor detects a spindle vibration frequency of 50 Hz. This multi-parameter collection provides a comprehensive foundation for subsequent analysis.

[0127] Step S420: Use the data preprocessing module to process the original collected data stream, denoise the cutting fluid flow value, cutting fluid temperature value and equipment vibration frequency data through the mean filtering method, and then unify the denoised data dimensions through the standardization method to obtain the preprocessed data stream.

[0128] In one possible implementation, the data preprocessing module optimizes the raw data stream. Mean filtering smoothes noise in flow, temperature, and vibration signals by taking the average of consecutive data. For example, flow data fluctuates between 178 and 182 liters / minute over 5 seconds, but stabilizes at 180 liters / minute after filtering.

[0129] Normalization methods unify data of varying dimensions into dimensionless values, making them easier for algorithms to process. For example, temperature data is converted from degrees Celsius to a standardized value on a scale of 0–1, ensuring it is on the same scale as flow and vibration data. This preprocessing improves data quality and lays the foundation for feature analysis.

[0130] Step S430: Perform feature analysis on the preprocessed data stream using a random forest algorithm, extract fluidity features based on the cutting fluid flow value, extract thermal stability features based on the cutting fluid temperature value, and extract equipment wear features based on the equipment vibration frequency data to obtain a feature vector set.

[0131] Specifically, the random forest algorithm preprocesses data streams through multiple decision tree analyses to extract key features. For flow values, the algorithm identifies fluidity characteristics, reflecting the circulation efficiency of the cutting fluid. For temperature values, it extracts thermal stability characteristics, indicating the fluid's ability to control temperature during machining. For vibration frequency, it extracts equipment wear characteristics, indicating potential damage to the spindle or tool. For example, in one analysis, the algorithm discovered that the flow rate was as low as 170 liters / minute and the fluidity characteristic value decreased. Combined with a high temperature of 50 degrees Celsius and a vibration frequency of 60 Hz, this indicated that the equipment might be wearing due to insufficient lubrication. This feature extraction provides precise input for fault prediction.

[0132] Step S440: Process the feature vector set through the pre-established fault prediction model, and use the logistic regression method to classify and predict the fluidity characteristics, thermal stability characteristics and equipment wear characteristics. If the prediction probability is higher than the preset threshold, it is determined that the equipment has a wear trend, and a wear trend prediction data stream is obtained.

[0133] Preferably, the fault prediction model is based on a logistic regression method and is classified by integrating fluidity, thermal stability and equipment wear characteristics.

[0134] The model has a preset probability threshold of 0.8. If the prediction result is 0.85, the equipment is considered to be experiencing wear. For example, after continuous operation, the eigenvector of a piece of equipment shows a fluidity eigenvalue of 0.6, a thermal stability eigenvalue of 0.7, and a wear eigenvalue of 0.9. The model outputs a wear probability of 0.87, triggering an alert. For example, a predicted data stream record shows a wear probability of 0.87 at 10:00:05 on May 3, 2025. This prediction mechanism quantifies risk and supports timely maintenance decisions.

[0135] It can be understood that the combination of the above methods forms a complete cutting fluid and equipment status monitoring link.

[0136] Sensor acquisition ensures comprehensive data, preprocessing improves data quality, feature analysis uncovers potential issues, and fault prediction provides actionable insights. For example, one factory used this system to proactively identify spindle wear risks, adjust cutting fluid formulations and maintenance plans, and avoid costly downtime. This interconnected design ensures stable equipment operation through multi-level analysis and precise prediction.

[0137] Furthermore, the fully automatic centralized cutting fluid supply control method provided in this embodiment includes step S500:

[0138] Step S510: Adjust the cutting fluid cycle parameters according to the wear trend prediction data stream.

[0139] Adjusting cutting fluid cycle parameters based on wear trend prediction data streams involves a deep integration of equipment status monitoring and cutting fluid management. The core is to dynamically optimize the fluid supply strategy to address equipment wear risks. For example, the wear trend prediction data stream typically contains key indicators of equipment operation, such as vibration frequency, cutting fluid flow rate, and temperature. Cycle parameters are adjusted by analyzing this data. For example, a processing equipment prediction data stream shows a wear probability of 0.87 at 2025-05-03 10:00:05, indicating increased spindle wear. At this point, the cycle needs to be shortened from the conventional 4-hour cycle to 2-hour cycle to enhance lubrication and cooling.

[0140] It should be noted that the cycle frequency is adjusted based on a wear probability threshold in the data stream. A higher threshold indicates a greater wear risk, and the cycle frequency needs to be increased accordingly. This dynamic adjustment mechanism ensures machining stability by responding to equipment status in real time.

[0141] Step S520: Determine whether the equipment wear is aggravated. If it is predicted that the equipment wear is aggravated, a linear regression method is used to optimize the liquid supply pressure setting value to obtain an adjusted purification frequency parameter and liquid supply pressure value.

[0142] In one possible implementation, if increased equipment wear is predicted, a linear regression method is used to optimize the fluid supply pressure setting.

[0143] Linear regression uses historical data to establish a relationship model between wear characteristics and fluid supply pressure, inputs the current wear characteristic value, and outputs the optimal pressure setting.

[0144] Specifically, during the operation of a certain factory equipment, the vibration frequency increased to 60 Hz, the cutting fluid flow rate dropped to 170 liters / minute, and the wear probability was 0.87.

[0145] A linear regression model analyzed historical data and determined that when the wear characteristic value reached 0.9, the fluid supply pressure needed to be increased from 2.5 bar to 3.0 bar. This adjusted pressure value enhanced the cutting fluid spray, improved tool lubrication, and slowed the wear process.

[0146] It is understandable that the optimization of the hydraulic pressure supply needs to be combined with the equipment load and the processing material. For example, when processing high-hardness alloys, the pressure adjustment range may be larger to ensure adequate coverage of the cutting fluid.

[0147] Preferably, the adjusted purge frequency parameter and the fluid supply pressure value need to be optimized in coordination to improve the cutting fluid performance.

[0148] Purification frequency refers to the interval between cutting fluid filtration and regeneration. As wear intensifies, metal chips and impurities accumulate more rapidly, necessitating increased purge frequency. For example, the standard purge frequency is once a day. When the wear probability exceeds 0.8, the frequency is adjusted to once every 12 hours. In one embodiment, after predicting wear trends for a particular device, the purge frequency is increased to twice a day, while the supply pressure is set at 3.0 bar and the flow rate is stabilized at 180 liters / minute. This adjustment improves cutting fluid cleanliness and reduces secondary wear on the tool caused by impurities.

[0149] It's important to note that increasing the frequency of cleaning needs to consider the capacity of the filter to avoid overloading. For example, one factory optimized cutting fluid management using the aforementioned method. To address the risk of spindle wear, they shortened the cycle to 2 hours, increased the fluid supply pressure to 3.0 bar, and adjusted the cleaning frequency to twice a day. Data streams showed a decrease in the wear probability from 0.87 to 0.65, demonstrating the effectiveness of the optimization measures. This multifaceted, coordinated adjustment, through comprehensive optimization of circulation, pressure, and cleaning, forms a robust equipment protection chain.

[0150] The adjustment of each parameter is based on the precise input of the predicted data flow, supporting each other to ensure the stability of equipment operation.

[0151] The present invention relates to a fully automatic centralized liquid supply control system for cutting fluid, which is used to implement the above-mentioned fully automatic centralized liquid supply control method for cutting fluid. The fully automatic centralized liquid supply control system for cutting fluid includes a first generation module, an adjustment module, a determination module, a second generation module and an update module, wherein the first generation module is used to obtain cutting fluid concentration data and liquid supply system pressure data through a sensor array, and use high-frequency sampling technology to generate real-time monitoring data. If the cutting fluid concentration data deviates from a preset threshold range or the liquid supply system pressure data exceeds a stable range, an abnormal signal is generated; the adjustment module is used to adjust the operating frequency of the liquid supply pump according to the abnormal signal using a proportional-integral-differential algorithm to form a closed-loop control system to maintain the target pressure value , and at the same time adjust the opening value of the liquid distribution valve according to the concentration deviation signal; the judgment module is used to extract the cutting fluid return from the adjusted liquid supply system, and process the cutting fluid return through a multi-stage filtration device and a centrifugal separation device. If it is detected that the impurity content of the cutting fluid return is lower than the preset threshold, it is determined that the purification treatment is completed; the second generation module is used to use a machine learning model to perform feature analysis on the performance parameters of the purified cutting fluid and the equipment operation data, and generate the equipment wear trend prediction result through a pre-established fault prediction model; the update module is used to adjust the circulation purification frequency and liquid supply parameters of the cutting fluid according to the equipment wear trend prediction result. If it is predicted that the equipment wear will increase, the purification frequency parameter is updated and the liquid supply pressure setting value is optimized.

[0152] Furthermore, the fully automatic centralized cutting fluid supply control system provided in this embodiment, the first generation module includes a first acquisition unit, a second acquisition unit, a third acquisition unit and a first generation unit, wherein the first acquisition unit is used to obtain cutting fluid concentration data and liquid supply system pressure data from the processing equipment through a sensor array using high-frequency sampling technology, generate a real-time monitoring data stream, and store it in a pre-established database to obtain continuous time series data; the second acquisition unit is used to compare the cutting fluid concentration data with a preset threshold range based on the continuous time series data, and if the cutting fluid concentration data exceeds the preset threshold range, it is marked as a concentration abnormality state, and a concentration abnormality mark is obtained; the third acquisition unit is used to compare the liquid supply system pressure data with a stable range based on the continuous time series data, and if the liquid supply system pressure data exceeds the stable range, it is marked as a pressure abnormality state, and a pressure abnormality mark is obtained; the first generation unit is used to obtain the concentration abnormality mark and the pressure abnormality mark through signal generation logic, and generate an abnormal signal if at least one of the concentration abnormality mark and the pressure abnormality mark is in an abnormal state.

[0153] Preferably, the fully automatic centralized cutting fluid supply control system provided in this embodiment, the adjustment module includes a second generation unit and a fourth acquisition unit, wherein the second generation unit is used to calculate the operating frequency adjustment amount of the liquid supply pump and the opening adjustment amount of the liquid distribution valve through a control algorithm according to the abnormal signal, and generate a frequency control instruction and an opening control instruction; the fourth acquisition unit is used to adjust the operating frequency of the liquid supply pump using the frequency control instruction, adjust the opening value of the liquid distribution valve through the opening control instruction, generate a closed-loop control data stream, store it in a database, and obtain a real-time adjustment record.

[0154] Furthermore, the fully automatic centralized cutting fluid supply control system provided in this embodiment has a judgment module including a fifth acquisition unit, a sixth acquisition unit, a seventh acquisition unit and an eighth acquisition unit, wherein the fifth acquisition unit is used to extract cutting fluid return from the liquid supply system through a return fluid collection device, and use a flow sensor to obtain flow data of the cutting fluid return to obtain a return fluid collection data stream; the sixth acquisition unit is used to process the cutting fluid return corresponding to the return fluid collection data stream through a multi-stage filtration device, use a first-stage filter to remove large particles of impurities, and then use a second-stage filter to remove tiny particles to obtain primary filtrate; the seventh acquisition unit is used to process the primary filtrate through a centrifugal separation device, use a centrifuge to separate residual particles and liquid at a constant speed, and obtain a secondary purified liquid; the eighth acquisition unit is used to obtain impurity content data of the secondary purified liquid through a detection sensor. If the impurity content is lower than a preset threshold, it is judged that the purification process is completed to obtain a purification completion data stream.

[0155] Compared with the existing technology, the fully automatic centralized cutting fluid supply control method and system provided in this embodiment monitors the cutting fluid concentration and supply pressure in real time through a sensor array, and automatically adjusts the supply pump frequency and the opening of the liquid distribution valve when an abnormality is detected, forming a closed-loop control system. At the same time, the recovered cutting fluid is subjected to multi-stage filtration and centrifugal separation purification treatment, and a machine learning model is used to analyze the cutting fluid performance parameters and equipment operation data to predict equipment wear trends. The cutting fluid circulation purification frequency and supply parameters are dynamically adjusted according to the prediction results to achieve intelligent management of the cutting fluid and preventive maintenance of equipment wear. This embodiment can maintain the stability of cutting fluid performance, extend the service life of equipment, improve processing quality and production efficiency, and has important practical value.

[0156] Although preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they are aware of the basic inventive concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the invention. Thus, the present invention is intended to include such changes and modifications as fall within the scope of the claims and their equivalents.

Claims

1. A fully automatic centralized cutting fluid supply control method, characterized in that: The following steps are involved: The sensor array acquires cutting fluid concentration data and fluid supply system pressure data, and uses high-frequency sampling technology to generate real-time monitoring data. If the cutting fluid concentration data deviates from a preset threshold range or the fluid supply system pressure data exceeds a stable range, an abnormal signal is generated; According to the abnormal signal, the proportional-integral-differential algorithm is used to adjust the operating frequency of the liquid supply pump to form a closed-loop control system to maintain the target pressure value, and at the same time, the opening value of the liquid distribution valve is adjusted according to the concentration deviation signal; Extracting cutting fluid return from the adjusted fluid supply system, processing the cutting fluid return through a multi-stage filtration device and a centrifugal separation device, and determining that the purification process is complete if the impurity content of the cutting fluid return is detected to be lower than a preset threshold; A machine learning model is used to analyze the characteristics of the purified cutting fluid performance parameters and equipment operation data, and a pre-established fault prediction model is used to generate equipment wear trend prediction results; The circulation purification frequency and fluid supply parameters of the cutting fluid are adjusted according to the equipment wear trend prediction results. If it is predicted that the equipment wear will increase, the purification frequency parameters are updated and the fluid supply pressure setting value is optimized.

2. The cutting fluid fully automatic centralized supply control method according to claim 1, characterized in that: The step of acquiring cutting fluid concentration data and fluid supply system pressure data through a sensor array and generating real-time monitoring data using a high-frequency sampling technique, and generating an abnormal signal if the cutting fluid concentration data deviates from a preset threshold range or the fluid supply system pressure data exceeds a stable range includes: By using high-frequency sampling technology through the sensor array, cutting fluid concentration data and fluid supply system pressure data are obtained from the processing equipment, generating real-time monitoring data streams, which are stored in a pre-established database to obtain continuous time series data; According to the continuous time series data, the cutting fluid concentration data is compared with a preset threshold range. If the cutting fluid concentration data exceeds the preset threshold range, it is marked as a concentration abnormal state, and a concentration abnormality flag is obtained; Comparing the pressure data of the liquid supply system with a stable range based on the continuous time series data, and if the pressure data of the liquid supply system exceeds the stable range, marking it as a pressure abnormality state, and obtaining a pressure abnormality flag; The concentration abnormality flag and the pressure abnormality flag are acquired through signal generation logic, and an abnormality signal is generated if at least one of the concentration abnormality flag and the pressure abnormality flag is in an abnormal state.

3. The cutting fluid fully automatic centralized supply control method according to claim 1, characterized in that: The steps of adjusting the operating frequency of the liquid supply pump using a proportional-integral-differential algorithm according to the abnormal signal to form a closed-loop control system to maintain the target pressure value, and adjusting the opening value of the liquid distribution valve according to the concentration deviation signal include: According to the abnormal signal, the operating frequency adjustment amount of the liquid supply pump and the opening adjustment amount of the liquid distribution valve are calculated by the control algorithm to generate a frequency control instruction and an opening control instruction; The frequency control instruction is used to adjust the operating frequency of the liquid supply pump, and the opening value of the liquid distribution valve is adjusted through the opening control instruction to generate a closed-loop control data stream, which is stored in a database to obtain real-time adjustment records.

4. The cutting fluid fully automatic centralized supply control method according to claim 1, characterized in that: The step of extracting the cutting fluid return from the adjusted fluid supply system, treating the cutting fluid return through a multi-stage filtration device and a centrifugal separation device, and determining that the purification process is complete if the impurity content of the cutting fluid return is detected to be lower than a preset threshold value comprises: Extracting cutting fluid return from the fluid supply system through a return fluid collection device, and using a flow sensor to obtain flow data of the cutting fluid return to obtain a return fluid collection data stream; Processing the cutting fluid return corresponding to the return fluid collection data stream through a multi-stage filtration device, using a first-stage filter to remove large particles of impurities, and then using a second-stage filter to remove small particles to obtain a primary filtrate; Processing the primary filtrate through a centrifugal separation device, using a centrifuge at a constant speed to separate residual particles from the liquid to obtain a secondary purified liquid; The impurity content data of the secondary purification liquid is obtained by detecting the sensor. If the impurity content is lower than a preset threshold, it is determined that the purification process is completed, and a purification completion data stream is obtained.

5. The cutting fluid fully automatic centralized supply control method according to claim 1, characterized in that: The steps of using a machine learning model to perform feature analysis on the purified cutting fluid performance parameters and equipment operation data, and generating equipment wear trend prediction results using a pre-established fault prediction model include: The cutting fluid performance parameters and equipment operation data are acquired from the equipment operation environment through the sensor data acquisition device. The flow sensor is used to collect the cutting fluid flow value, the temperature sensor is used to collect the cutting fluid temperature value, and the vibration sensor is used to collect the equipment vibration frequency data to obtain the original collected data stream; A data preprocessing module is used to process the raw collected data stream, and a mean filtering method is used to perform denoising on the cutting fluid flow value, the cutting fluid temperature value, and the equipment vibration frequency data. The denoised data is then dimensionally unified using a standardization method to obtain a preprocessed data stream; Performing feature analysis on the pre-processed data stream using a random forest algorithm, extracting fluidity features based on cutting fluid flow values, extracting thermal stability features based on cutting fluid temperature values, and extracting equipment wear features based on equipment vibration frequency data, to obtain a feature vector set; The feature vector set is processed by a pre-established fault prediction model, and the fluidity characteristics, thermal stability characteristics and equipment wear characteristics are classified and predicted using a logistic regression method. If the prediction probability is higher than a preset threshold, it is determined that the equipment has a wear trend, and a wear trend prediction data stream is obtained.

6. The cutting fluid fully automatic centralized supply control method according to claim 1, characterized in that: The steps of adjusting the circulation purification frequency and fluid supply parameters of the cutting fluid according to the equipment wear trend prediction result, and updating the purification frequency parameters and optimizing the fluid supply pressure setting value if the equipment wear is predicted to be aggravated, include: Adjust cutting fluid cycle parameters based on wear trend prediction data stream; Determine whether the equipment wear is aggravated. If it is predicted that the equipment wear is aggravated, the linear regression method is used to optimize the liquid supply pressure setting value to obtain the adjusted purification frequency parameters and liquid supply pressure value.

7. A cutting fluid fully automatic centralized supply control system, used to implement the cutting fluid fully automatic centralized supply control method according to any one of claims 1 to 6, characterized in that: The cutting fluid fully automatic centralized supply control system includes: A first generation module is configured to acquire cutting fluid concentration data and fluid supply system pressure data through a sensor array, generate real-time monitoring data using a high-frequency sampling technique, and generate an abnormal signal if the cutting fluid concentration data deviates from a preset threshold range or the fluid supply system pressure data exceeds a stable range; an adjustment module for adjusting the operating frequency of the liquid supply pump using a proportional-integral-differential algorithm according to the abnormal signal, forming a closed-loop control system to maintain the target pressure value, and adjusting the opening value of the liquid distribution valve according to the concentration deviation signal; a determination module, configured to extract cutting fluid return from the adjusted fluid supply system, process the cutting fluid return through a multi-stage filtration device and a centrifugal separation device, and determine that purification processing is complete if it is detected that the impurity content of the cutting fluid return is lower than a preset threshold; The second generation module is used to use a machine learning model to perform feature analysis on the performance parameters of the purified cutting fluid and the equipment operation data, and to generate equipment wear trend prediction results through a pre-established fault prediction model; The updating module is used to adjust the circulation purification frequency and fluid supply parameters of the cutting fluid according to the equipment wear trend prediction result. If it is predicted that the equipment wear will increase, the purification frequency parameter is updated and the fluid supply pressure setting value is optimized.

8. The fully automatic centralized cutting fluid supply control system according to claim 7, characterized in that: The first generation module includes: The first acquisition unit is used to acquire cutting fluid concentration data and fluid supply system pressure data from the processing equipment through a sensor array using high-frequency sampling technology, generate a real-time monitoring data stream, and store it in a pre-established database to obtain continuous time series data; a second acquiring unit, configured to compare the cutting fluid concentration data with a preset threshold range based on the continuous time series data, and mark the cutting fluid concentration data as an abnormal concentration state if the cutting fluid concentration data exceeds the preset threshold range, thereby obtaining an abnormal concentration flag; a third acquiring unit, configured to compare the pressure data of the liquid supply system with a stable range based on the continuous time series data, and if the pressure data of the liquid supply system exceeds the stable range, mark it as a pressure abnormality state, thereby obtaining a pressure abnormality flag; The first generating unit is configured to obtain the concentration abnormality flag and the pressure abnormality flag through signal generation logic, and generate an abnormal signal if at least one of the concentration abnormality flag and the pressure abnormality flag is in an abnormal state.

9. The fully automatic centralized cutting fluid supply control system according to claim 7, characterized in that: The adjustment module includes: a second generating unit, configured to calculate, according to the abnormal signal, an operating frequency adjustment amount of the liquid supply pump and an opening adjustment amount of the liquid distribution valve through a control algorithm, and generate a frequency control instruction and an opening control instruction; The fourth acquisition unit is used to adjust the operating frequency of the liquid supply pump using the frequency control instruction, adjust the opening value of the liquid distribution valve through the opening control instruction, generate a closed-loop control data stream, store it in a database, and obtain a real-time adjustment record.

10. The fully automatic centralized cutting fluid supply control system according to claim 7, characterized in that: The determination module includes: A fifth acquisition unit is configured to extract cutting fluid return from the fluid supply system through a return fluid acquisition device, and acquire flow rate data of the cutting fluid return using a flow sensor to obtain a return fluid acquisition data stream; a sixth acquisition unit, configured to process the cutting fluid return corresponding to the return fluid collection data stream through a multi-stage filtration device, using a first-stage filter to remove large particles of impurities, and then using a second-stage filter to remove small particles, to obtain a primary filtrate; a seventh obtaining unit, configured to process the primary filtrate through a centrifugal separation device, using a centrifuge at a constant speed to separate residual particles from the liquid to obtain a secondary purified liquid; The eighth acquisition unit is used to acquire the impurity content data of the secondary purification liquid through a detection sensor. If the impurity content is lower than a preset threshold, it is determined that the purification process is completed, and a purification completion data stream is obtained.

Citation Information

Patent Citations

  • Bi-GRU network-based tool remaining service life prediction method under different working conditions

    CN115186571A

  • Cutting fluid control system and method in numerical control machining based on machine learning

    CN118348914A

  • Precision control method and device of linear cutting machine and readable medium

    CN118635606A

  • Electromagnetic valve accurate control method based on flow dynamic adjustment

    CN119244805A

  • Numerical control tool machining monitoring method

    CN119828596A

Cited By

  • Automatic control system for soybean milk separation

    CN120754612A

  • Integrated online measurement system for thermal physical property parameters of cutting fluid

    CN121007935A