Purpose. To substantiate the application of the artificial neural network predictors in the water-supply pump control systems for forecasting the water neural network predictor have been obtained by experimental research studies on the operating water supply system. The mathematical model of the electric drive system of a pump has been created by using the object-oriented method. This model takes into consideration the nonlinearity of magnetic core characteristics of an induction motor and existence of the hydraulic network with distributed parameters at the pump exit. Using computer simulation, the following aspects were researched: the frequency regulation mode of asynchronous drive of a pump, dynamic ratings of pump drive for predicted value of water consumption and quality performance of pressure regulation. Findings. On the base of experimentally obtained data the artificial neural network predictor, which adequately forecasts the water consumption value within 24 hours, has been synthesized. The systems controlling an artificial neural network predictor pump have been synthesized; they realize different laws of the pump output pressure regulation. The laws of frequency control of water-supply pump electric drive system have been substantiated. Originality. For the first time, it is proposed to use the artificial neural network predictor in the pump control system to forecast the value of water consumption and synthesis of the control influences on the pump electric drive with taking the water consumption into consideration. Results, which demonstrate the mutual influences of processes in the water-supply network and electromagnetic processes in the electric drive with induction motor, have been obtained. Practical value. Due to consideration of the predicted value of water consumption in the offered pump control system, the efficacy of the water supply system has increased while the unproductive losses of water have decreased.

 Наукова періодика України 

---

Додаткова інформація про автора(ів) публікації:
(cписок формується автоматично, до списку можуть бути включені персоналії з подібними іменами або однофамільці)

• Гнатушенко Володимир Володимирович (1975–) (технічні науки)
• Васильєв Всеволод Вікторович (1935–2021) (технічні науки)

  Якщо, ви не знайшли інформацію про автора(ів) публікації, маєте бажання виправити або відобразити більш докладну інформацію про науковців України запрошуємо заповнити "Анкету науковця"