Drastic change in crop prices is observed due to climatic changes, natural calamities and lack of quantity of a specific commodity. Crop price prediction plays key role in effective farm management. Farmers are not able to predict these crop prices and facing massive loss. These aspects pressure us to use advanced technology and develop accurate, reliable and efficient crop price prediction system. Crop price prediction also helps agriculture based industries and policymakers. There are many price-sensitive crops like tomatoes, onions, potatoes, Soybean and other food grains, which need prior price prediction so that farmers can take wise decisions on which crop to cultivate. Functional link neural network is chosen to develop Basic network for Soybeans price prediction. Optimization algorithms like whale optimization, particle swarm optimization and Harris Hawks Optimization are used to calculate appropriate biases and weights. Dataset is taken from daily reports issued by Chicago Mercantile Exchange (CME).Most efficient hybrid FLNN with associated Expansion function, activation function and learning scheme for predicting crop price could be found out through our study.